Vaguery + hey-i-know-this-guy   82

 « earlier
[1904.08658] Batch Tournament Selection for Genetic Programming
Lexicase selection achieves very good solution quality by introducing ordered test cases. However, the computational complexity of lexicase selection can prohibit its use in many applications. In this paper, we introduce Batch Tournament Selection (BTS), a hybrid of tournament and lexicase selection which is approximately one order of magnitude faster than lexicase selection while achieving a competitive quality of solutions. Tests on a number of regression datasets show that BTS compares well with lexicase selection in terms of mean absolute error while having a speed-up of up to 25 times. Surprisingly, BTS and lexicase selection have almost no difference in both diversity and performance. This reveals that batches and ordered test cases are completely different mechanisms which share the same general principle fostering the specialization of individuals. This work introduces an efficient algorithm that sheds light onto the main principles behind the success of lexicase, potentially opening up a new range of possibilities for algorithms to come.
genetic-programming  selection  algorithms  hey-I-know-this-guy  (I-wish-it-wasn't-always-a-guy)  to-write-about  to-do
april 2019 by Vaguery
[1810.07800] Alignments as Compositional Structures
Alignments, i.e., position-wise comparisons of two or more strings or ordered lists are of utmost practical importance in computational biology and a host of other fields, including historical linguistics and emerging areas of research in the Digital Humanities. The problem is well-known to be computationally hard as soon as the number of input strings is not bounded. Due to its prac- tical importance, a huge number of heuristics have been devised, which have proved very successful in a wide range of applications. Alignments nevertheless have received hardly any attention as formal, mathematical structures. Here, we focus on the compositional aspects of alignments, which underlie most algo- rithmic approaches to computing alignments. We also show that the concepts naturally generalize to finite partially ordered sets and partial maps between them that in some sense preserve the partial orders.
discrete-mathematics  optimization  alignments  combinatorics  hey-I-know-this-guy  bioinformatics  rather-interesting  formalization  to-write-about  consider:multiobjective-optimization  consider:fitness-landscapes  question:transitivity
april 2019 by Vaguery
Geometric semantic genetic programming for recursive boolean programs
Geometric Semantic Genetic Programming (GSGP) induces a unimodal fitness landscape for any problem that consists in finding a function fitting given input/output examples. Most of the work around GSGP to date has focused on real-world applications and on improving the originally proposed search operators, rather than on broadening its theoretical framework to new domains. We extend GSGP to recursive programs, a notoriously challenging domain with highly discontinuous fitness landscapes. We focus on programs that map variable-length Boolean lists to Boolean values, and design search operators that are provably efficient in the training phase and attain perfect generalization. Computational experiments complement the theory and demonstrate the superiority of the new operators to the conventional ones. This work provides new insights into the relations between program syntax and semantics, search operators and fitness landscapes, also for more general recursive domains.
april 2019 by Vaguery
[1903.07008] Leveling the Playing Field -- Fairness in AI Versus Human Game Benchmarks
From the beginning if the history of AI, there has been interest in games as a platform of research. As the field developed, human-level competence in complex games became a target researchers worked to reach. Only relatively recently has this target been finally met for traditional tabletop games such as Backgammon, Chess and Go. Current research focus has shifted to electronic games, which provide unique challenges. As is often the case with AI research, these results are liable to be exaggerated or misrepresented by either authors or third parties. The extent to which these games benchmark consist of fair competition between human and AI is also a matter of debate. In this work, we review the statements made by authors and third parties in the general media and academic circle about these game benchmark results and discuss factors that can impact the perception of fairness in the contest between humans and machines
engineering-criticism  rather-interesting  hey-I-know-this-guy  performance-measure  what-gets-measured-gets-fudged  artificial-intelligence  games  machine-learning  to-write-about  benchmarking
april 2019 by Vaguery
The Emergence of Canalization and Evolvability in an Open-Ended, Interactive Evolutionary System | Artificial Life | MIT Press Journals
Many believe that an essential component for the discovery of the tremendous diversity in natural organisms was the evolution of evolvability, whereby evolution speeds up its ability to innovate by generating a more adaptive pool of offspring. One hypothesized mechanism for evolvability is developmental canalization, wherein certain dimensions of variation become more likely to be traversed and others are prevented from being explored (e.g., offspring tend to have similar-size legs, and mutations affect the length of both legs, not each leg individually). While ubiquitous in nature, canalization is rarely reported in computational simulations of evolution, which deprives us of in silico examples of canalization to study and raises the question of which conditions give rise to this form of evolvability. Answering this question would shed light on why such evolvability emerged naturally, and it could accelerate engineering efforts to harness evolution to solve important engineering challenges. In this article, we reveal a unique system in which canalization did emerge in computational evolution. We document that genomes entrench certain dimensions of variation that were frequently explored during their evolutionary history. The genetic representation of these organisms also evolved to be more modular and hierarchical than expected by chance, and we show that these organizational properties correlate with increased fitness. Interestingly, the type of computational evolutionary experiment that produced this evolvability was very different from traditional digital evolution in that there was no objective, suggesting that open-ended, divergent evolutionary processes may be necessary for the evolution of evolvability.
artificial-life  contingency  evolution  theoretical-biology  evolvability  to-write-about  hey-I-know-this-guy
march 2019 by Vaguery
[1711.08477] Benchmarking Relief-Based Feature Selection Methods for Bioinformatics Data Mining
Modern biomedical data mining requires feature selection methods that can (1) be applied to large scale feature spaces (e.g. omics' data), (2) function in noisy problems, (3) detect complex patterns of association (e.g. gene-gene interactions), (4) be flexibly adapted to various problem domains and data types (e.g. genetic variants, gene expression, and clinical data) and (5) are computationally tractable. To that end, this work examines a set of filter-style feature selection algorithms inspired by the Relief' algorithm, i.e. Relief-Based algorithms (RBAs). We implement and expand these RBAs in an open source framework called ReBATE (Relief-Based Algorithm Training Environment). We apply a comprehensive genetic simulation study comparing existing RBAs, a proposed RBA called MultiSURF, and other established feature selection methods, over a variety of problems. The results of this study (1) support the assertion that RBAs are particularly flexible, efficient, and powerful feature selection methods that differentiate relevant features having univariate, multivariate, epistatic, or heterogeneous associations, (2) confirm the efficacy of expansions for classification vs. regression, discrete vs. continuous features, missing data, multiple classes, or class imbalance, (3) identify previously unknown limitations of specific RBAs, and (4) suggest that while MultiSURF* performs best for explicitly identifying pure 2-way interactions, MultiSURF yields the most reliable feature selection performance across a wide range of problem types.
machine-learning  bioinformatics  hey-I-know-this-guy  feature-selection  benchmarking  epistasis  algorithms  to-write-about
february 2019 by Vaguery
[1812.05225] Finding the origin of noise transients in LIGO data with machine learning
Quality improvement of interferometric data collected by gravitational-wave detectors such as Advanced LIGO and Virgo is mission critical for the success of gravitational-wave astrophysics. Gravitational-wave detectors are sensitive to a variety of disturbances of non-astrophysical origin with characteristic frequencies in the instrument band of sensitivity. Removing non-astrophysical artifacts that corrupt the data stream is crucial for increasing the number and statistical significance of gravitational-wave detections and enabling refined astrophysical interpretations of the data. Machine learning has proved to be a powerful tool for analysis of massive quantities of complex data in astronomy and related fields of study. We present two machine learning methods, based on random forest and genetic programming algorithms, that can be used to determine the origin of non-astrophysical transients in the LIGO detectors. We use two classes of transients with known instrumental origin that were identified during the first observing run of Advanced LIGO to show that the algorithms can successfully identify the origin of non-astrophysical transients in real interferometric data and thus assist in the mitigation of instrumental and environmental disturbances in gravitational-wave searches. While the data sets described in this paper are specific to LIGO, and the exact procedures employed were unique to the same, the random forest and genetic programming code bases and means by which they were applied as a dual machine learning approach are completely portable to any number of instruments in which noise is believed to be generated through mechanical couplings, the source of which is not yet discovered.
genetic-programming  hey-I-know-this-guy  astrophysics  data-analysis  data-mining  to-understand  feature-construction  classification
january 2019 by Vaguery
What else is in an evolved name? Exploring evolvable specificity with SignalGP [PeerJ Preprints]
Tags are evolvable labels that provide genetic programs a flexible mechanism for specification. Tags are used to label and refer to programmatic elements, such as functions or jump targets. However, tags differ from traditional, more rigid methods for handling labeling because they allow for inexact references; that is, a referring tag need not exactly match its referent. Here, we explore how adjusting the threshold for how what qualifies as a match affects adaptive evolution. Further, we propose broadened applications of tags in the context of a genetic programming (GP) technique called SignalGP. SignalGP gives evolution direct access to the event-driven paradigm. Program modules in SignalGP are tagged and can be triggered by signals (with matching tags) from the environment, from other agents, or due to internal regulation. Specifically, we propose to extend this tag based system to: (1) provide more fine-grained control over module execution and regulation (e.g., promotion and repression) akin to natural gene regulatory networks, (2) employ a mosaic of GP representations within a single program, and (3) facilitate major evolutionary transitions in individuality (i.e., allow hierarchical program organization to evolve de novo).
artificial-life  genetic-programming  representation  hey-I-know-this-guy  the-mangle-in-practice  to-examine  consider:ReQ
january 2019 by Vaguery
Accelerating open source LLVM development - Software Tools blog - Software Tools - Arm Community
We are currently using Works on Arm as the underlying platform to run these build jobs. Works on Arm provides free-of-charge Arm-based infrastructure for open source projects. In that way, we gain access to powerful build servers to complete such a challenging task within acceptable build time constraints.
ARM  to-write-about  hey-I-know-this-guy  opensource  resources  to-do
january 2019 by Vaguery
[quant-ph/0208149] A semi-quantum version of the game of Life
A version of John Conway's game of Life is presented where the normal binary values of the cells are replaced by oscillators which can represent a superposition of states. The original game of Life is reproduced in the classical limit, but in general additional properties not seen in the original game are present that display some of the effects of a quantum mechanical Life. In particular, interference effects are seen.
quantums  Game-of-Life  hey-I-know-this-guy  cellular-automata
november 2018 by Vaguery
Semantic information, agency, & physics | Interface Focus
Shannon information theory provides various measures of so-called syntactic information, which reflect the amount of statistical correlation between systems. By contrast, the concept of ‘semantic information’ refers to those correlations which carry significance or ‘meaning’ for a given system. Semantic information plays an important role in many fields, including biology, cognitive science and philosophy, and there has been a long-standing interest in formulating a broadly applicable and formal theory of semantic information. In this paper, we introduce such a theory. We define semantic information as the syntactic information that a physical system has about its environment which is causally necessary for the system to maintain its own existence. ‘Causal necessity’ is defined in terms of counter-factual interventions which scramble correlations between the system and its environment, while ‘maintaining existence’ is defined in terms of the system's ability to keep itself in a low entropy state. We also use recent results in non-equilibrium statistical physics to analyse semantic information from a thermodynamic point of view. Our framework is grounded in the intrinsic dynamics of a system coupled to an environment, and is applicable to any physical system, living or otherwise. It leads to formal definitions of several concepts that have been intuitively understood to be related to semantic information, including ‘value of information’, ‘semantic content’ and ‘agency’.
complexity  philosophy-of-science  information-theory  define-your-terms  hey-I-know-this-guy  semantics  to-understand  cannot-read
november 2018 by Vaguery
Adam Kotsko The Political Theology of Neoliberalism - state of nature
Neoliberals do rely on libertarian rhetoric, but libertarianism is basically neoliberalism for fools. When neoliberals are talking amongst themselves, they always acknowledge that a strong state is absolutely necessary to their agenda. This is because markets do not spontaneously arise in the absence of state interference, or in other words, markets are not natural. They must be artificially constructed, and so one way of defining neoliberalism is as a project to use state power to cultivate or create markets so that people will be forced to be free in the neoliberal sense.
neoliberalism  interview  quotes  hey-I-know-this-guy  to-write-about  fascism  political-economy  financial-crisis  capitalism  worldview
october 2018 by Vaguery
[1806.01387] New And Surprising Ways to Be Mean. Adversarial NPCs with Coupled Empowerment Minimisation
Creating Non-Player Characters (NPCs) that can react robustly to unforeseen player behaviour or novel game content is difficult and time-consuming. This hinders the design of believable characters, and the inclusion of NPCs in games that rely heavily on procedural content generation. We have previously addressed this challenge by means of empowerment, a model of intrinsic motivation, and demonstrated how a coupled empowerment maximisation (CEM) policy can yield generic, companion-like behaviour. In this paper, we extend the CEM framework with a minimisation policy to give rise to adversarial behaviour. We conduct a qualitative, exploratory study in a dungeon-crawler game, demonstrating that CEM can exploit the affordances of different content facets in adaptive adversarial behaviour without modifications to the policy. Changes to the level design, underlying mechanics and our character's actions do not threaten our NPC's robustness, but yield new and surprising ways to be mean.
hey-I-know-this-guy  coevolution  evolutionary-algorithms  engineering-design  rather-interesting  to-write-about
june 2018 by Vaguery
[1806.02717] Gamorithm
Examining games from a fresh perspective we present the idea of game-inspired and game-based algorithms, dubbed "gamorithms".
june 2018 by Vaguery
Home | Kappa Language
By separating a rule from a patch on which it acts we gain a much clearer approach to mechanistic causality. If causal analysis were to proceed at the level of patches, it would obfuscate the causal structure of a system by dragging along context irrelevant to an event. In addition to simulation and static analysis, the Kappa platform also extracts the causal structure of a rule system from its simulation traces.
bioinformatics  representation  hey-I-know-this-guy  complexology  pattern-discovery  rather-interesting  to-write-about
may 2018 by Vaguery
Trent McConaghy - FFX
FFX is a technique for symbolic regression, to induce whitebox models given X/y training data. It does Fast Function Extraction. It is:

Fast - runtime 5-60 seconds, depending on problem size (1GHz cpu)
Scalable - 1000 input variables, no problem!
Deterministic - no need to "hope and pray".
If you ignore the whitebox-model aspect, FFX can be viewed as a regression tool. It's been used this way for thousands of industrial problems with 100K+ input variables. It can also be used as a classifier (FFXC), by wrapping the output with a logistic map. This has also been used successfully on thousands of industrial problems.
hey-I-know-this-guy  symbolic-regression  algorithms  numerical-methods  data-analysis  to-write-about
may 2018 by Vaguery
[1804.05445] Evolving Event-driven Programs with SignalGP
We present SignalGP, a new genetic programming (GP) technique designed to incorporate the event-driven programming paradigm into computational evolution's toolbox. Event-driven programming is a software design philosophy that simplifies the development of reactive programs by automatically triggering program modules (event-handlers) in response to external events, such as signals from the environment or messages from other programs. SignalGP incorporates these concepts by extending existing tag-based referencing techniques into an event-driven context. Both events and functions are labeled with evolvable tags; when an event occurs, the function with the closest matching tag is triggered. In this work, we apply SignalGP in the context of linear GP. We demonstrate the value of the event-driven paradigm using two distinct test problems (an environment coordination problem and a distributed leader election problem) by comparing SignalGP to variants that are otherwise identical, but must actively use sensors to process events or messages. In each of these problems, rapid interaction with the environment or other agents is critical for maximizing fitness. We also discuss ways in which SignalGP can be generalized beyond our linear GP implementation.
may 2018 by Vaguery
[1803.05859v3] Neural Network Quine
Self-replication is a key aspect of biological life that has been largely overlooked in Artificial Intelligence systems. Here we describe how to build and train self-replicating neural networks. The network replicates itself by learning to output its own weights. The network is designed using a loss function that can be optimized with either gradient-based or non-gradient-based methods. We also describe a method we call regeneration to train the network without explicit optimization, by injecting the network with predictions of its own parameters. The best solution for a self-replicating network was found by alternating between regeneration and optimization steps. Finally, we describe a design for a self-replicating neural network that can solve an auxiliary task such as MNIST image classification. We observe that there is a trade-off between the network's ability to classify images and its ability to replicate, but training is biased towards increasing its specialization at image classification at the expense of replication. This is analogous to the trade-off between reproduction and other tasks observed in nature. We suggest that a self-replication mechanism for artificial intelligence is useful because it introduces the possibility of continual improvement through natural selection.
artificial-life  machine-learning  quines  rather-interesting  to-write-about  hey-I-know-this-guy
may 2018 by Vaguery
Complexity Explorer
New for 2018, our flagship course, Introduction to Complexity, will be open year round. All units will be available at all times, so you can learn the fundamentals of Complex Systems Science at your own pace, and earn your certificate at any time.

All other courses are designed to be similar to a semester-long course, offering the chance to delve into a subject deeply. Courses are available in two ways: as a course session, or an archived course.  Each course is usually offered in one session per calendar year.

Tutorials are short, self-paced “mini-courses” designed to introduce students to important techniques and to provide illustrations of their application in complex systems.

Complexity Explorer's courses and tutorials are supported by user donations and contributions from the Santa Fe Institute.  Please consider donating to support further course and tutorial development.
april 2018 by Vaguery
[1803.10122] World Models
We explore building generative neural network models of popular reinforcement learning environments. Our world model can be trained quickly in an unsupervised manner to learn a compressed spatial and temporal representation of the environment. By using features extracted from the world model as inputs to an agent, we can train a very compact and simple policy that can solve the required task. We can even train our agent entirely inside of its own hallucinated dream generated by its world model, and transfer this policy back into the actual environment.
An interactive version of this paper is available at this https URL
hey-I-know-this-guy  machine-learning  representation  agents  data-fusion  algorithms  generative-models
march 2018 by Vaguery
Dave Snowden at TedX: A succinct overview of his groundbreaking work | More Beyond
In less than 18 minutes, Dave manages to introduce complex systems theory; tell the children’s party story (3 mins 30 secs) and introduce a new theory of change based on the power of micro-narrative and vector measures enabled by Sensemaker (7 mins).
cynefin  TED-talk  emergent-design  engineering-philosophy  public-policy  hey-I-know-this-guy
february 2018 by Vaguery
World Wide Wanderings
Over the years I have worked on many different research and computing projects, all over the world. Most of my scientific work is related to the origin of life, evolution, and complex systems & emergence. Below is a brief overview of the main projects that I am currently working on. Preprints of my publications on this work are available for download (see the list of publications).
february 2018 by Vaguery
Information Theory of Complex Networks
Complex networks are characterized by highly heterogeneous distributions of links, often pervading the presence of key properties such as robustness under node removal. Several correlation measures have been defined in order to characterize the structure of these nets. Here we show that mutual information, noise and joint entropies can be properly defined on a static graph. These measures are computed for a number of real networks and analytically estimated for some simple standard models. It is shown that real networks are clustered in a well-defined domain of the entropy- noise space. By using simulated annealing optimization, it is shown that optimally heterogeneous nets actually cluster around the same narrow domain, suggesting that strong constraints actually operate on the possible universe of complex networks. The evolutionary implications are discussed
december 2017 by Vaguery
[1711.07387] How morphological development can guide evolution
Organisms result from multiple adaptive processes occurring and interacting at different time scales. One such interaction is that between development and evolution. In modeling studies, it has been shown that development sweeps over a series of traits in a single agent, and sometimes exposes promising static traits. Subsequent evolution can then canalize these rare traits. Thus, development can, under the right conditions, increase evolvability. Here, we report on a previously unknown phenomenon when embodied agents are allowed to develop and evolve: Evolution discovers body plans which are robust to control changes, these body plans become genetically assimilated, yet controllers for these agents are not assimilated. This allows evolution to continue climbing fitness gradients by tinkering with the developmental programs for controllers within these permissive body plans. This exposes a previously unknown detail about the Baldwin effect: instead of all useful traits becoming genetically assimilated, only phenotypic traits that render the agent robust to changes in other traits become assimilated. We refer to this phenomenon as differential canalization. This finding also has important implications for the evolutionary design of artificial and embodied agents such as robots: robots that are robust to internal changes in their controllers may also be robust to external changes in their environment, such as transferal from simulation to reality, or deployment in novel environments.
artificial-life  evolved-devo  developmental-biology  representation  rather-interesting  genetic-programming  hey-I-know-this-guy
december 2017 by Vaguery
[1705.00594] A System for Accessible Artificial Intelligence
While artificial intelligence (AI) has become widespread, many commercial AI systems are not yet accessible to individual researchers nor the general public due to the deep knowledge of the systems required to use them. We believe that AI has matured to the point where it should be an accessible technology for everyone. We present an ongoing project whose ultimate goal is to deliver an open source, user-friendly AI system that is specialized for machine learning analysis of complex data in the biomedical and health care domains. We discuss how genetic programming can aid in this endeavor, and highlight specific examples where genetic programming has automated machine learning analyses in previous projects.
hey-I-know-this-guy  user-experience  machine-learning  visualization  user-interface  one-ring  to-write-about  to-go-see
september 2017 by Vaguery
[1709.05915] Push and Pull Search for Solving Constrained Multi-objective Optimization Problems
This paper proposes a push and pull search (PPS) framework for solving constrained multi-objective optimization problems (CMOPs). To be more specific, the proposed PPS divides the search process into two different stages, including the push and pull search stages. In the push stage, a multi-objective evolutionary algorithm (MOEA) is adopted to explore the search space without considering any constraints, which can help to get across infeasible regions very fast and approach the unconstrained Pareto front. Furthermore, the landscape of CMOPs with constraints can be probed and estimated in the push stage, which can be utilized to conduct the parameters setting for constraint-handling approaches applied in the pull stage. Then, a constrained multi-objective evolutionary algorithm (CMOEA) equipped with an improved epsilon constraint-handling is applied to pull the infeasible individuals achieved in the push stage to the feasible and non-dominated regions. Compared with other CMOEAs, the proposed PPS method can more efficiently get across infeasible regions and converge to the feasible and non-dominated regions by applying push and pull search strategies at different stages. To evaluate the performance regarding convergence and diversity, a set of benchmark CMOPs is used to test the proposed PPS and compare with other five CMOEAs, including MOEA/D-CDP, MOEA/D-SR, C-MOEA/D, MOEA/D-Epsilon and MOEA/D-IEpsilon. The comprehensive experimental results demonstrate that the proposed PPS achieves significantly better or competitive performance than the other five CMOEAs on most of the benchmark set.
hey-I-know-this-guy  multiobjective-optimization  metaheuristics  evolutionary-algorithms  algorithms  to-write-about  nudge-targets  consider:implementing
september 2017 by Vaguery
[1708.03157] TensorFlow Enabled Genetic Programming
Genetic Programming, a kind of evolutionary computation and machine learning algorithm, is shown to benefit significantly from the application of vectorized data and the TensorFlow numerical computation library on both CPU and GPU architectures. The open source, Python Karoo GP is employed for a series of 190 tests across 6 platforms, with real-world datasets ranging from 18 to 5.5M data points. This body of tests demonstrates that datasets measured in tens and hundreds of data points see 2-15x improvement when moving from the scalar/SymPy configuration to the vector/TensorFlow configuration, with a single core performing on par or better than multiple CPU cores and GPUs. A dataset composed of 90,000 data points demonstrates a single vector/TensorFlow CPU core performing 875x better than 40 scalar/Sympy CPU cores. And a dataset containing 5.5M data points sees GPU configurations out-performing CPU configurations on average by 1.3x.
hey-I-know-this-guy  genetic-programming  symbolic-regression  library  GPU  to-write-about
september 2017 by Vaguery
[1504.04909] Illuminating search spaces by mapping elites
Many fields use search algorithms, which automatically explore a search space to find high-performing solutions: chemists search through the space of molecules to discover new drugs; engineers search for stronger, cheaper, safer designs, scientists search for models that best explain data, etc. The goal of search algorithms has traditionally been to return the single highest-performing solution in a search space. Here we describe a new, fundamentally different type of algorithm that is more useful because it provides a holistic view of how high-performing solutions are distributed throughout a search space. It creates a map of high-performing solutions at each point in a space defined by dimensions of variation that a user gets to choose. This Multi-dimensional Archive of Phenotypic Elites (MAP-Elites) algorithm illuminates search spaces, allowing researchers to understand how interesting attributes of solutions combine to affect performance, either positively or, equally of interest, negatively. For example, a drug company may wish to understand how performance changes as the size of molecules and their cost-to-produce vary. MAP-Elites produces a large diversity of high-performing, yet qualitatively different solutions, which can be more helpful than a single, high-performing solution. Interestingly, because MAP-Elites explores more of the search space, it also tends to find a better overall solution than state-of-the-art search algorithms. We demonstrate the benefits of this new algorithm in three different problem domains ranging from producing modular neural networks to designing simulated and real soft robots. Because MAP- Elites (1) illuminates the relationship between performance and dimensions of interest in solutions, (2) returns a set of high-performing, yet diverse solutions, and (3) improves finding a single, best solution, it will advance science and engineering.
september 2017 by Vaguery
Tools for Diagnosing Domain Name Issues · Chris Salzman's Website
Oftentimes when inheriting a web project you also inherit myriad domain name related issues. And, as always, the command line is your friend for finding quick information associated with that domain name. Here’s a few tools I’ve been using a lot lately:
hey-I-know-this-guy  web-design  devops  to-learn
september 2017 by Vaguery
CSS Grid is Good · Chris Salzman's Website
Ease of coding is one thing, but grid also makes it far easier to separate the semantic markup of a page from the visual styling. Your templates don’t need to be strewn about with div’s to make it all work. Instead you piece the page together semantically and then visually it can match your…well, your vision.
august 2017 by Vaguery
[1708.05070] Data-driven Advice for Applying Machine Learning to Bioinformatics Problems
As the bioinformatics field grows, it must keep pace not only with new data but with new algorithms. Here we contribute a thorough analysis of 13 state-of-the-art, commonly used machine learning algorithms on a set of 165 publicly available classification problems in order to provide data-driven algorithm recommendations to current researchers. We present a number of statistical and visual comparisons of algorithm performance and quantify the effect of model selection and algorithm tuning for each algorithm and dataset. The analysis culminates in the recommendation of five algorithms with hyperparameters that maximize classifier performance across the tested problems, as well as general guidelines for applying machine learning to supervised classification problems.
meta-optimization  machine-learning  benchmarking  performance-measure  feature-construction  to-write-about  hey-I-know-this-guy
august 2017 by Vaguery
[1605.09304] Synthesizing the preferred inputs for neurons in neural networks via deep generator networks
Deep neural networks (DNNs) have demonstrated state-of-the-art results on many pattern recognition tasks, especially vision classification problems. Understanding the inner workings of such computational brains is both fascinating basic science that is interesting in its own right - similar to why we study the human brain - and will enable researchers to further improve DNNs. One path to understanding how a neural network functions internally is to study what each of its neurons has learned to detect. One such method is called activation maximization (AM), which synthesizes an input (e.g. an image) that highly activates a neuron. Here we dramatically improve the qualitative state of the art of activation maximization by harnessing a powerful, learned prior: a deep generator network (DGN). The algorithm (1) generates qualitatively state-of-the-art synthetic images that look almost real, (2) reveals the features learned by each neuron in an interpretable way, (3) generalizes well to new datasets and somewhat well to different network architectures without requiring the prior to be relearned, and (4) can be considered as a high-quality generative method (in this case, by generating novel, creative, interesting, recognizable images).
hey-I-know-this-guy  neural-networks  generative-models  machine-learning  GPTP  nudge-targets  to-write-about
august 2017 by Vaguery
[1705.08971] Optimal Cooperative Inference
Cooperative transmission of data fosters rapid accumulation of knowledge by efficiently combining experience across learners. Although well studied in human learning, there has been less attention to cooperative transmission of data in machine learning, and we consequently lack strong formal frameworks through which we may reason about the benefits and limitations of cooperative inference. We present such a framework. We introduce a novel index for measuring the effectiveness of probabilistic information transmission, and cooperative information transmission specifically. We relate our cooperative index to previous measures of teaching in deterministic settings. We prove conditions under which optimal cooperative inference can be achieved, including a representation theorem which constrains the form of inductive biases for learners optimized for cooperative inference. We conclude by demonstrating how these principles may inform the design of machine learning algorithms and discuss implications for human learning, machine learning, and human-machine learning systems.
hey-I-know-this-guy  machine-learning  relevance-theory  philosophy  rather-interesting  to-write-about  pedagogy  communication-and-learning
august 2017 by Vaguery
[1706.04119] From MEGATON to RASCAL: Surfing the Parameter Space of Evolutionary Algorithms
The practice of evolutionary algorithms involves a mundane yet inescapable phase, namely, finding parameters that work well. How big should the population be? How many generations should the algorithm run? What is the (tournament selection) tournament size? What probabilities should one assign to crossover and mutation? All these nagging questions need good answers if one is to embrace success. Through an extensive series of experiments over multiple evolutionary algorithm implementations and problems we show that parameter space tends to be rife with viable parameters. We aver that this renders the life of the practitioner that much easier, and cap off our study with an advisory digest for the weary.
hey-I-know-this-guy  machine-learning  meta-optimization  evolutionary-algorithms  metaheuristics  hyperheuristics  to-write-about
june 2017 by Vaguery
Does aligning phenotypic and genotypic modularity improve the evolution of neural networks? | Evolving AI Lab
Many argue that to evolve artificial intelligence that rivals that of natural animals, we need to evolve neural networks that are structurally organized in that they exhibit modularity, regularity, and hierarchy. It was recently shown that a cost for network connections, which encourages the evolution of modularity, can be combined with an indirect encoding, which encourages the evolution of regularity, to evolve networks that are both modular and regular. However, the bias towards regularity from indirect encodings may prevent evolution from independently optimizing different modules to perform different functions, unless modularity in the phenotype is aligned with modularity in the genotype. We test this hypothesis on two multi-modal problems—a pattern recognition task and a robotics task—that each require different phenotypic modules. In general, we find that performance is improved only when genotypic and phenotypic modularity are encouraged simultaneously, though the role of alignment remains unclear. In addition, intuitive manual decompositions fail to provide the performance benefits of automatic methods on the more challenging robotics problem, emphasizing the importance of automatic, rather than manual, decomposition methods. These results suggest encouraging modularity in both the genotype and phenotype as an important step towards solving large-scale multi-modal problems, but also indicate that more research is required before we can evolve structurally organized networks to solve tasks that require multiple, different neural modules.
hey-I-know-this-guy  evolutionary-algorithms  performance-measure  metrics  to-write-about  to-do
june 2017 by Vaguery
BIRDS-Lab
Professor Revzen and his team at the Biologically Inspired Robotics and Dynamical Systems (BIRDS) Lab are working on discovering, modeling, and reproducing the strategies animals use when interacting with physical objects. This work consists of collaboration with biomechanists to analyze experimental data, developing new mathematical tools for modeling and estimation of model parameters, and construction of robots which employ the new principles.
hey-I-know-this-guy  robotics  engineering-design  engineering  nudge-targets  to-write-about  local
may 2017 by Vaguery
[1703.00512] PMLB: A Large Benchmark Suite for Machine Learning Evaluation and Comparison
The selection, development, or comparison of machine learning methods in data mining can be a difficult task based on the target problem and goals of a particular study. Numerous publicly available real-world and simulated benchmark datasets have emerged from different sources, but their organization and adoption as standards have been inconsistent. As such, selecting and curating specific benchmarks remains an unnecessary burden on machine learning practitioners and data scientists. The present study introduces an accessible, curated, and developing public benchmark resource to facilitate identification of the strengths and weaknesses of different machine learning methodologies. We compare meta-features among the current set of benchmark datasets in this resource to characterize the diversity of available data. Finally, we apply a number of established machine learning methods to the entire benchmark suite and analyze how datasets and algorithms cluster in terms of performance. This work is an important first step towards understanding the limitations of popular benchmarking suites and developing a resource that connects existing benchmarking standards to more diverse and efficient standards in the future.
hey-I-know-this-guy  machine-learning  benchmarking  horse-races  performance-measure  to-write-about
april 2017 by Vaguery
[1702.01780] Toward the automated analysis of complex diseases in genome-wide association studies using genetic programming
Machine learning has been gaining traction in recent years to meet the demand for tools that can efficiently analyze and make sense of the ever-growing databases of biomedical data in health care systems around the world. However, effectively using machine learning methods requires considerable domain expertise, which can be a barrier of entry for bioinformaticians new to computational data science methods. Therefore, off-the-shelf tools that make machine learning more accessible can prove invaluable for bioinformaticians. To this end, we have developed an open source pipeline optimization tool (TPOT-MDR) that uses genetic programming to automatically design machine learning pipelines for bioinformatics studies. In TPOT-MDR, we implement Multifactor Dimensionality Reduction (MDR) as a feature construction method for modeling higher-order feature interactions, and combine it with a new expert knowledge-guided feature selector for large biomedical data sets. We demonstrate TPOT-MDR's capabilities using a combination of simulated and real world data sets from human genetics and find that TPOT-MDR significantly outperforms modern machine learning methods such as logistic regression and eXtreme Gradient Boosting (XGBoost). We further analyze the best pipeline discovered by TPOT-MDR for a real world problem and highlight TPOT-MDR's ability to produce a high-accuracy solution that is also easily interpretable.
hey-I-know-this-guy  bioinformatics  machine-learning  meta-optimization  workflows  framework
april 2017 by Vaguery
[1306.5667] Using Genetic Programming to Model Software
We study a generic program to investigate the scope for automatically customising it for a vital current task, which was not considered when it was first written. In detail, we show genetic programming (GP) can evolve models of aspects of BLAST's output when it is used to map Solexa Next-Gen DNA sequences to the human genome.
bioinformatics  software-synthesis  algorithms  genetic-programming  hey-I-know-this-guy  nudge-targets  consider:looking-to-see
april 2017 by Vaguery
[1704.05143] The Emergence of Canalization and Evolvability in an Open-Ended, Interactive Evolutionary System
Natural evolution has produced a tremendous diversity of functional organisms. Many believe an essential component of this process was the evolution of evolvability, whereby evolution speeds up its ability to innovate by generating a more adaptive pool of offspring. One hypothesized mechanism for evolvability is developmental canalization, wherein certain dimensions of variation become more likely to be traversed and others are prevented from being explored (e.g. offspring tend to have similarly sized legs, and mutations affect the length of both legs, not each leg individually). While ubiquitous in nature, canalization almost never evolves in computational simulations of evolution. Not only does that deprive us of in silico models in which to study the evolution of evolvability, but it also raises the question of which conditions give rise to this form of evolvability. Answering this question would shed light on why such evolvability emerged naturally and could accelerate engineering efforts to harness evolution to solve important engineering challenges. In this paper we reveal a unique system in which canalization did emerge in computational evolution. We document that genomes entrench certain dimensions of variation that were frequently explored during their evolutionary history. The genetic representation of these organisms also evolved to be highly modular and hierarchical, and we show that these organizational properties correlate with increased fitness. Interestingly, the type of computational evolutionary experiment that produced this evolvability was very different from traditional digital evolution in that there was no objective, suggesting that open-ended, divergent evolutionary processes may be necessary for the evolution of evolvability.
evolutionary-algorithms  diversity  novelty  cause-and-effect  rather-interesting  to-write-about  hey-I-know-this-guy  nudge-targets  consider:similar-analyses
april 2017 by Vaguery
[1106.5531] Balls in Boxes: Variations on a Theme of Warren Ewens and Herbert Wilf
We comment on, elaborate, and extend the work of Warren Ewens and Herbert Wilf, described in their this http URL about the maximum in balls-and-boxes problem. In particular we meta-apply their ingenious method to show that it is not really needed, and that one is better off using the so-called Poisson Approximation, at least in applications to the real world, because extremely unlikely events mever happen in real life. This article is accompanied by the Maple package this http URL">BallsInBoxes.
probability-theory  statistics  hey-I-know-this-guy  combinatorics  looking-to-see  define-your-terms  to-write-about  mathematical-recreations  nudge-targets  consider:rediscovery
april 2017 by Vaguery
[1109.4994] The finite-state character of physical dynamics
Finite physical systems have only a finite amount of distinct state. This finiteness is fundamental in statistical mechanics, where the maximum number of distinct states compatible with macroscopic constraints defines entropy. Here we show that finiteness of distinct state is similarly fundamental in ordinary mechanics: energy and momentum are defined by the maximum number of distinct states possible in a given time or distance. More generally, any moment of energy or momentum bounds distinct states in time or space. These results generalise both the Nyquist bandwidth-bound on distinct values in classical signals, and quantum uncertainty bounds. The new certainty bounds are achieved by finite-bandwidth evolutions in which time and space are effectively discrete, including quantum evolutions that are effectively classical. Since energy and momentum count distinct states, they are defined in classical finite-state dynamics, and they relate classical relativity to finite-state evolution.
hey-I-know-this-guy  physics  automata  theoretical-physics  complexology  quantum  discrete-mathematics  philosophy-of-science  representation
march 2017 by Vaguery
[1501.01300] Minimum Probabilistic Finite State Learning Problem on Finite Data Sets: Complexity, Solution and Approximations
In this paper, we study the problem of determining a minimum state probabilistic finite state machine capable of generating statistically identical symbol sequences to samples provided. This problem is qualitatively similar to the classical Hidden Markov Model problem and has been studied from a practical point of view in several works beginning with the work presented in: Shalizi, C.R., Shalizi, K.L., Crutchfield, J.P. (2002) \textit{An algorithm for pattern discovery in time series.} Technical Report 02-10-060, Santa Fe Institute. arxiv.org/abs/cs.LG/0210025. We show that the underlying problem is NP-hard and thus all existing polynomial time algorithms must be approximations on finite data sets. Using our NP-hardness proof, we show how to construct a provably correct algorithm for constructing a minimum state probabilistic finite state machine given data and empirically study its running time.
hey-I-know-this-guy  automata  time-series  optimization  modeling  performance-measure  computational-complexity  rather-interesting  to-write-about  nudge-targets  consider:rewriting-the-damned-algorithm
march 2017 by Vaguery
[1305.2537] On Creativity of Elementary Cellular Automata
We map cell-state transition rules of elementary cellular automata (ECA) onto the cognitive control versus schizotypy spectrum phase space and interpret cellular automaton behaviour in terms of creativity. To implement the mapping we draw analogies between a degree of schizotypy and generative diversity of ECA rules, and between cognitive control and robustness of ECA rules (expressed via Derrida coefficient). We found that null and fixed point ECA rules lie in the autistic domain and chaotic rules are 'schizophrenic'. There are no highly articulated 'creative' ECA rules. Rules closest to 'creativity' domains are two-cycle rules exhibiting wave-like patterns in the space-time evolution.
hey-I-know-this-guy  cellular-automata  emergent-design  complexology  stamp-collecting  to-write-about  consider:feature-discovery
march 2017 by Vaguery
[1409.3588] Density Classification Quality of the Traffic-majority Rules
The density classification task is a famous problem in the theory of cellular automata. It is unsolvable for deterministic automata, but recently solutions for stochastic cellular automata have been found. One of them is a set of stochastic transition rules depending on a parameter η, the traffic-majority rules.
Here I derive a simplified model for these cellular automata. It is valid for a subset of the initial configurations and uses random walks and generating functions. I compare its prediction with computer simulations and show that it expresses recognition quality and time correctly for a large range of η values.
cellular-automata  hey-I-know-this-guy  emergent-design  to-write-about  nudge-targets  consider:representation  consider:higher-order-problems  consider:classification-of-rules
march 2017 by Vaguery
Why Facts Don’t Change Our Minds - The New Yorker
If reason is designed to generate sound judgments, then it’s hard to conceive of a more serious design flaw than confirmation bias. Imagine, Mercier and Sperber suggest, a mouse that thinks the way we do. Such a mouse, “bent on confirming its belief that there are no cats around,” would soon be dinner. To the extent that confirmation bias leads people to dismiss evidence of new or underappreciated threats—the human equivalent of the cat around the corner—it’s a trait that should have been selected against. The fact that both we and it survive, Mercier and Sperber argue, proves that it must have some adaptive function, and that function, they maintain, is related to our “hypersociability.”

Mercier and Sperber prefer the term “myside bias.” Humans, they point out, aren’t randomly credulous. Presented with someone else’s argument, we’re quite adept at spotting the weaknesses. Almost invariably, the positions we’re blind about are our own.
hey-I-know-this-guy  Sperberism  rather-interesting  anthropology  cognition  to-write-about  salient
march 2017 by Vaguery
[1701.09097] An Intermediate Level of Abstraction for Computational Systems Chemistry
Computational techniques are required for narrowing down the vast space of possibilities to plausible prebiotic scenarios, since precise information on the molecular composition, the dominant reaction chemistry, and the conditions for that era are scarce. The exploration of large chemical reaction networks is a central aspect in this endeavour. While quantum chemical methods can accurately predict the structures and reactivities of small molecules, they are not efficient enough to cope with large-scale reaction systems. The formalization of chemical reactions as graph grammars provides a generative system, well grounded in category theory, at the right level of abstraction for the analysis of large and complex reaction networks. An extension of the basic formalism into the realm of integer hyperflows allows for the identification of complex reaction patterns, such as auto-catalysis, in large reaction networks using optimization techniques.
cheminformatics  graph-theory  representation  network-theory  hey-I-know-this-guy  complexology
february 2017 by Vaguery
[1612.02537] Another phase transition in the Axelrod model
Axelrod's model of cultural dissemination, despite its apparent simplicity, demonstrates complex behavior that has been of much interest in statistical physics. Despite the many variations and extensions of the model that have been investigated, a systematic investigation of the effects of changing the size of the neighborhood on the lattice in which interactions can occur has not been made. Here we investigate the effect of varying the radius R of the von Neumann neighborhood in which agents can interact. We show, in addition to the well-known phase transition at the critical value of q, the number of traits, another phase transition at a critical value of R, and draw a q -- R phase diagram for the Axelrod model on a square lattice. In addition, we present a mean-field approximation of the model in which behavior on an infinite lattice can be analyzed.
hey-I-know-this-guy  evolutionary-economics  artificial-life  simulation  phase-transitions  to-write-about
february 2017 by Vaguery
Existence and construction of large stable food webs | bioRxiv
Ecological diversity is ubiquitous despite the restrictions imposed by competitive exclusion and apparent competition. To explain the observed richness of species in a given habitat, food web theory has explored nonlinear functional responses, self-interaction or spatial structure and dispersal - model ingredients that have proven to promote stability and diversity. We here instead return to classical Lotka-Volterra equations, where species-species interaction is characterized by a simple product and spatial restrictions are ignored. We quantify how this idealization imposes constraints on coexistence and diversity for many species. To this end, we introduce the concept of free and controlled species and use this to demonstrate how stable food webs can be constructed by sequential addition of species. When we augment the resulting network by additional weak interactions we are able to show that it is possible to construct large food webs of arbitrary connectivity. Our model thus serves as a formal starting point for the study of sustainable interaction patterns between species.
hey-I-know-this-guy  theoretical-biology  food-webs  ecology  rather-interesting  biological-engineering  dynamical-systems  nudge-targets  consider:feature-discovery
february 2017 by Vaguery
[1510.08697] Systems poised to criticality through Pareto selective forces
Pareto selective forces optimize several targets at the same time, instead of single fitness functions. Systems subjected to these forces evolve towards their Pareto front, a geometrical object akin to the thermodynamical Gibbs surface and whose shape and differential geometry underlie the existence of phase transitions. In this paper we outline the connection between the Pareto front and criticality and critical phase transitions. It is shown how, under definite circumstances, Pareto selective forces drive a system towards a critical ensemble that separates the two phases of a first order phase transition. Different mechanisms implementing such Pareto selective dynamics are revised.
hey-I-know-this-guy  multiobjective-optimization  fitness-landscapes  my-thesis-stuff  theoretical-biology  complexology  small-world
january 2017 by Vaguery
[1606.07772] The emotional arcs of stories are dominated by six basic shapes
Advances in computing power, natural language processing, and digitization of text now make it possible to study our a culture's evolution through its texts using a "big data" lens. Our ability to communicate relies in part upon a shared emotional experience, with stories often following distinct emotional trajectories, forming patterns that are meaningful to us. Here, by classifying the emotional arcs for a filtered subset of 1,737 stories from Project Gutenberg's fiction collection, we find a set of six core trajectories which form the building blocks of complex narratives. We strengthen our findings by separately applying optimization, linear decomposition, supervised learning, and unsupervised learning. For each of these six core emotional arcs, we examine the closest characteristic stories in publication today and find that particular emotional arcs enjoy greater success, as measured by downloads.
via:cshalizi  hey-I-know-this-guy  complexology  digital-humanities  storytelling  feature-construction  rather-interesting  nudge-targets  consider:feature-discovery
july 2016 by Vaguery
[1603.09419] The free energy requirements of biological organisms; implications for evolution
Recent advances in nonequilibrium statistical physics have provided unprecedented insight into the thermodynamics of dynamic processes. The author recently used these advances to extend Landauer's semi-formal reasoning concerning the thermodynamics of bit erasure, to derive the minimal free energy required to implement an arbitrary computation. Here, I extend this analysis, deriving the minimal free energy required by an organism to run a given (stochastic) map π from its sensor inputs to its actuator outputs. I use this result to calculate the input-output map π of an organism that optimally trades off the free energy needed to run π with the phenotypic fitness that results from implementing π. I end with a general discussion of the limits imposed on the rate of the terrestrial biosphere's information processing by the flux of sunlight on the Earth.
july 2016 by Vaguery
Entropy | Free Full-Text | The Free Energy Requirements of Biological Organisms; Implications for Evolution
Recent advances in nonequilibrium statistical physics have provided unprecedented insight into the thermodynamics of dynamic processes. The author recently used these advances to extend Landauer’s semi-formal reasoning concerning the thermodynamics of bit erasure, to derive the minimal free energy required to implement an arbitrary computation. Here, I extend this analysis, deriving the minimal free energy required by an organism to run a given (stochastic) map π from its sensor inputs to its actuator outputs. I use this result to calculate the input-output map π of an organism that optimally trades off the free energy needed to run π with the phenotypic fitness that results from implementing π. I end with a general discussion of the limits imposed on the rate of the terrestrial biosphere’s information processing by the flux of sunlight on the Earth.
hey-I-know-this-guy  theoretical-biology  complexology  systems-biology  physics  information-theory
april 2016 by Vaguery
http://arxiv.org/abs/1603.08233
Pattern recognition and classification is a central concern for modern information processing systems. In particular, one key challenge to image and video classification has been that the computational cost of image processing scales linearly with the number of pixels in the image or video. Here we present an intelligent machine (the "active categorical classifier," or ACC) that is inspired by the saccadic movements of the eye, and is capable of classifying images by selectively scanning only a portion of the image. We harness evolutionary computation to optimize the ACC on the MNIST hand-written digit classification task, and provide a proof-of-concept that the ACC works on noisy multi-class data. We further analyze the ACC and demonstrate its ability to classify images after viewing only a fraction of the pixels, and provide insight on future research paths to further improve upon the ACC presented here.
hey-I-know-this-guy  machine-learning  lexicase-selection  sampling  approximation  nudge-targets  consider:relation-to-data-balancing  consider:relation-to-lexicase
april 2016 by Vaguery
[1601.02918] Spatial self-organization in hybrid models of multicellular adhesion
Spatial self-organization emerges in distributed systems exhibiting local interactions when nonlinearities and the appropriate propagation of signals are at work. These kinds of phenomena can be modeled with different frameworks, typically cellular automata or reaction-diffusion systems. A different class of dynamical processes involves the correlated movement of agents over space, which can be mediated through chemotactic movement or minimization of cell-cell interaction energy. A classic example of the latter is given by the formation of spatially segregated assemblies when cells display differential adhesion. Here we consider a new class of dynamical models, involving cell adhesion among two stochastically exchangeable cell states as a minimal model capable of exhibiting well-defined, ordered spatial patterns. Our results suggest that a whole space of pattern-forming rules is hosted by the combination of physical differential adhesion and the value of probabilities modulating cell phenotypic switching, showing that Turing-like patterns can be obtained without resorting to reaction-diffusion processes. If the model is expanded allowing cells to proliferate and die in an environment where diffusible nutrient and toxic waste are at play, different phases are observed, characterized by regularly spaced patterns. The analysis of the parameter space reveals that certain phases reach higher population levels than other modes of organization. A detailed exploration of the mean-field theory is also presented. Finally we let populations of cells with different adhesion matrices compete for reproduction, showing that, in our model, structural organization can improve the fitness of a given cell population. The implications of these results for ecological and evolutionary models of pattern formation and the emergence of multicellularity are outlined.
cellular-automata  pattern-formation  rather-interesting  theoretical-biology  hey-I-know-this-guy  good-to-see-somebody-finally-do-this  nudge-targets  consider:writing-up
march 2016 by Vaguery
[1603.02481] A Software Package for Chemically Inspired Graph Transformation
Chemical reaction networks can be automatically generated from graph grammar descriptions, where rewrite rules model reaction patterns. Because a molecule graph is connected and reactions in general involve multiple molecules, the rewriting must be performed on multisets of graphs. We present a general software package for this type of graph rewriting system, which can be used for modelling chemical systems. The package contains a C++ library with algorithms for working with transformation rules in the Double Pushout formalism, e.g., composition of rules and a domain specific language for programming graph language generation. A Python interface makes these features easily accessible. The package also has extensive procedures for automatically visualising not only graphs and rewrite rules, but also Double Pushout diagrams and graph languages in form of directed hypergraphs. The software is available as an open source package, and interactive examples can be found on the accompanying webpage.
reaction-networks  complexology  hey-I-know-this-guy  self-organization  simulation  software  nudge-targets  consider:exploring
march 2016 by Vaguery
[1510.08697] Systems poised to criticality through Pareto selective forces
Pareto selective forces optimise several targets at the same time, instead of single fitness functions. Systems subjected to these forces evolve towards their Pareto front, a geometrical object akin to the thermodynamical Gibbs surface and whose shape and differential geometry underlie the existence of phase transitions. In this paper we outline the connection of the Pareto front with criticality and critical phase transitions. It is shown how, under definite circumstances, Pareto selective forces drive a system towards a critical ensemble that separates the two phases of a first order phase transition. Different mechanisms implementing such Pareto selective dynamics are revised.
multiobjective-optimization  hey-I-know-this-guy  took-long-enough  fitness-landscapes  nudge-targets  to-write-about
february 2016 by Vaguery
[1601.00900] Too good to be true: when overwhelming evidence fails to convince
Is it possible for a large sequence of measurements or observations, which support a hypothesis, to counterintuitively decrease our confidence? Can unanimous support be too good to be true? The assumption of independence is often made in good faith, however rarely is consideration given to whether a systemic failure has occurred.
Taking this into account can cause certainty in a hypothesis to decrease as the evidence for it becomes apparently stronger. We perform a probabilistic Bayesian analysis of this effect with examples based on (i) archaeological evidence, (ii) weighing of legal evidence, and (iii) cryptographic primality testing.
We find that even with surprisingly low systemic failure rates high confidence is very difficult to achieve and in particular we find that certain analyses of cryptographically-important numerical tests are highly optimistic, underestimating their false-negative rate by as much as a factor of 280.
via:arsyed  probability-theory  hey-I-know-this-guy  statistics  decision-making  paradox  rather-interesting  nudge-targets  consider:exploration-vs-exploitation
january 2016 by Vaguery
[1511.02079] Emergence of proto-organisms from bistable stochastic differentiation and adhesion
The rise of multicellularity in the early evolution of life represents a major challenge for evolutionary biology. Guidance for finding answers has emerged from disparate fields, from phylogenetics to modelling and synthetic biology, but little is known about the potential origins of multicellular aggregates before genetic programs took full control of developmental processes. Such aggregates should involve spatial organisation of differentiated cells and the modification of flows and concentrations of metabolites within well defined boundaries. Here we show that, in an environment where limited nutrients and toxic metabolites are introduced, a population of cells capable of stochastic differentiation and differential adhesion can develop into multicellular aggregates with a complex internal structure. The morphospace of possible patterns is shown to be very rich, including proto-organisms that display a high degree of organisational complexity, far beyond simple heterogeneous populations of cells. Our findings reveal that there is a potentially enormous richness of organismal complexity between simple mixed cooperators and embodied living organisms.
artificial-life  hey-I-know-this-guy  origin-of-life  simulation  nudge-targets  consider:exploration
december 2015 by Vaguery
[1401.1942] Test Problem Construction for Single-Objective Bilevel Optimization
In this paper, we propose a procedure for designing controlled test problems for single-objective bilevel optimization. The construction procedure is flexible and allows its user to control the different complexities that are to be included in the test problems independently of each other. In addition to properties that control the difficulty in convergence, the procedure also allows the user to introduce difficulties caused by interaction of the two levels. As a companion to the test problem construction framework, the paper presents a standard test suite of twelve problems, which includes eight unconstrained and four constrained problems. Most of the problems are scalable in terms of variables and constraints. To provide baseline results, we have solved the proposed test problems using a nested bilevel evolutionary algorithm. The results can be used for comparison, while evaluating the performance of any other bilevel optimization algorithm. The codes related to the paper may be accessed from the website \url{this http URL}.
bilevel-optimization  multiobjective-optimization  hey-I-know-this-guy  algorithms  operations-research  performance-measure  rather-interesting  representation  nudge-targets  consider:representation
november 2015 by Vaguery
Replaying Evolution to Test the Cause of Extinction of One Ecotype in an Experimentally Evolved Population | bioRxiv
In a long-term evolution experiment with Escherichia coli, bacteria in one of twelve populations evolved the ability to consume citrate, a previously unexploited resource in a glucose-limited medium. This innovation led to the frequency-dependent coexistence of citrate-consuming (Cit+) and non-consuming (Cit–) ecotypes, with Cit– bacteria persisting on the exogenously supplied glucose as well as other carbon molecules released by the Cit+ bacteria. After more than 10,000 generations of coexistence, however, the Cit– lineage went extinct; cells with the Cit– phenotype dropped to levels below detection, and the Cit– clade could not be detected by molecular assays based on its unique genotype. We hypothesized that this extinction event was a deterministic outcome of evolutionary change within the population, specifically the appearance of a more-fit Cit+ ecotype that competitively excluded the Cit– ecotype. We tested this hypothesis by re-evolving the population from one frozen sample taken just prior to the extinction and from another sample taken several thousand generations earlier, in each case for 500 generations and with 20-fold replication. To our surprise, the Cit– type did not go extinct in any of these replays, and Cit– cells also persisted in a single replicate that was propagated for 3,000 generations. Even more unexpectedly, we showed that the Cit– ecotype could reinvade the Cit+ population after its extinction. Taken together, these results indicate that the extinction of the Cit– ecotype was not a deterministic outcome driven by competitive exclusion by the Cit+ ecotype. The extinction also cannot be explained by demographic stochasticity, as the population size of the Cit– ecotype should have been many thousands of cells even during the daily transfer events. Instead, we infer that the extinction must have been caused by a rare chance event in which some aspect of the experimental conditions was inadvertently perturbed.
evolutionary-biology  experiment  hey-I-know-this-guy  microbiology  real-life-Fontana-experiments
september 2015 by Vaguery
Sustained fitness gains and variability in fitness trajectories in the long-term evolution experiment with Escherichia coli | bioRxiv
Many populations live in environments subject to frequent biotic and abiotic changes. Nonetheless, it is interesting to ask whether an evolving population's mean fitness can increase indefinitely, and potentially without any limit, even in a constant environment. A recent study showed that fitness trajectories of Escherichia coli populations over 50,000 generations were better described by a power-law model than by a hyperbolic model. According to the power-law model, the rate of fitness gain declines over time but fitness has no upper limit, whereas the hyperbolic model implies a hard limit. Here, we examine whether the previously estimated power-law model predicts the fitness trajectory for an additional 10,000 generations. To that end, we conducted more than 1100 new competitive fitness assays. Consistent with the previous study, the power-law model fits the new data better than the hyperbolic model. We also analysed the variability in fitness among populations, finding subtle, but significant, heterogeneity in mean fitness. Some, but not all, of this variation reflects differences in mutation rate that evolved over time. Taken together, our results imply that both adaptation and divergence can continue indefinitely-or at least for a long time-even in a constant environment.
evolutionary-biology  experiment  hey-I-know-this-guy  biology  looking-to-see
september 2015 by Vaguery
Understanding Society: A survey of agent-based models
Federico Bianchi and Flaminio Squazzoni have published a very useful survey of the development and uses of agent-based models in the social sciences over the past twenty-five years in WIREs Comput Stat 2015 (link). The article is a very useful reference and discussion for anyone interested in the applicability of ABM within sociology.
hey-I-know-this-guy  agent-based  review  rather-interesting  visualization  evolutionary-economics
september 2015 by Vaguery
[1504.01434] The X-rule: universal computation in a non-isotropic Life-like Cellular Automaton
We present a new Life-like cellular automaton (CA) capable of logic universality -- the X-rule. The CA is 2D, binary, with a Moore neighborhood and λ parameter similar to the game-of-Life, but is not based on birth/survival and is non-isotropic. We outline the search method. Several glider types and stable structures emerge spontaneously within X-rule dynamics. We construct glider-guns based on periodic oscillations between stable barriers, and interactions to create logical gates.
cellular-automata  hey-I-know-this-guy  complexology  rather-interesting  nudge-targets  consider:feature-discovery
september 2015 by Vaguery
[1502.00809] The consensus in the two-feature two-state one-dimensional Axelrod model revisited
The Axelrod model for the dissemination of culture exhibits a rich spatial distribution of cultural domains, which depends on the values of the two model parameters: F, the number of cultural features and q, the common number of states each feature can assume. In the one-dimensional model with F=q=2, which is closely related to the constrained voter model, Monte Carlo simulations indicate the existence of multicultural absorbing configurations in which at least one macroscopic domain coexist with a multitude of microscopic ones in the thermodynamic limit. However, rigorous analytical results for the infinite system starting from the configuration where all cultures are equally likely show convergence to only monocultural or consensus configurations. Here we show that this disagreement is due simply to the order that the time-asymptotic limit and the thermodynamic limit are taken in the simulations. In addition, we show how the consensus-only result can be derived using Monte Carlo simulations of finite chains.
evolutionary-economics  hey-I-know-this-guy  agent-based  economics  cultural-dynamics  simulation  artificial-life  to-read
september 2015 by Vaguery
[1505.01887] Optimal Neuron Selection: NK Echo State Networks for Reinforcement Learning
This paper introduces the NK Echo State Network. The problem of learning in the NK Echo State Network is reduced to the problem of optimizing a special form of a Spin Glass Problem known as an NK Landscape. No weight adjustment is used; all learning is accomplished by spinning up (turning on) or spinning down (turning off) neurons in order to find a combination of neurons that work together to achieve the desired computation. For special types of NK Landscapes, an exact global solution can be obtained in polynomial time using dynamic programming. The NK Echo State Network is applied to a reinforcement learning problem requiring a recurrent network: balancing two poles on a cart given no velocity information. Empirical results shows that the NK Echo State Network learns very rapidly and yields very good generalization.
hey-I-know-this-guy  Nk-landscapes  Kauffmania  spin-glasses  neural-networks  rather-interesting  machine-learning  nudge-targets  consider:representation
august 2015 by Vaguery
[1501.04497] Food web assembly rules
In food webs, many interacting species coexist despite the restrictions imposed by the competitive exclusion principle and apparent competition. For the generalized Lotka-Volterra equations, sustainable coexistence necessitates nonzero determinant of the interaction matrix. Here we show that this requirement is equivalent to demanding that each species be part of a non-overlapping pairing, which substantially constrains the food web structure. We demonstrate that a stable food web can always be obtained if a non-overlapping pairing exists. If it does not, the matrix rank can be used to quantify the lack of niches, corresponding to unpaired species. For the species richness at each trophic level, we derive the food web assembly rules, which specify sustainable combinations. In neighboring levels, these rules allow the higher level to avert competitive exclusion at the lower, thereby incorporating apparent competition. In agreement with data, the assembly rules predict high species numbers at intermediate levels and thinning at the top and bottom. Using comprehensive food web data, we demonstrate how omnivores or parasites with hosts at multiple trophic levels can loosen the constraints and help obtain coexistence in food webs. Hence, omnivory may be the glue that keeps communities intact even under extinction or ecological release of species.
hey-I-know-this-guy  food-webs  community-assembly  theoretical-biology  population-biology  ecology  simulation  self-organization  nudge-targets  rather-interesting
july 2015 by Vaguery
[1210.7019] Labyrinthine clustering in a spatial rock-paper-scissors ecosystem
The spatial rock-paper-scissors ecosystem, where three species interact cyclically, is a model example of how spatial structure can maintain biodiversity. We here consider such a system for a broad range of interaction rates. When one species grows very slowly, this species and its prey dominate the system by self-organizing into a labyrinthine configuration in which the third species propagates. The cluster size distributions of the two dominating species have heavy tails and the configuration is stabilized through a complex, spatial feedback loop. We introduce a new statistical measure that quantifies the amount of clustering in the spatial system by comparison with its mean field approximation. Hereby, we are able to quantitatively explain how the labyrinthine configuration slows down the dynamics and stabilizes the system.
rock-paper-scissors  roshambo  artificial-life  agent-based  self-organization  pattern-formation  hey-I-know-this-guy  nudge-targets
july 2015 by Vaguery
[1402.4466] Compressed bitmap indexes: beyond unions and intersections
Compressed bitmap indexes are used to speed up simple aggregate queries in databases. Indeed, set operations like intersections, unions and complements can be represented as logical operations (AND,OR,NOT) that are ideally suited for bitmaps. However, it is less obvious how to apply bitmaps to more advanced queries. For example, we might seek products in a store that meet some, but maybe not all, criteria. Such threshold queries generalize intersections and unions; they are often used in information-retrieval and data-mining applications. We introduce new algorithms that are sometimes three orders of magnitude faster than a naive approach. Our work shows that bitmap indexes are more broadly applicable than is commonly believed.
algorithms  optimization  database  rather-interesting  hey-I-know-this-guy  nudge-targets  consider:rediscovery
july 2015 by Vaguery
Composing Music with LSTM Recurrent Networks
Here are some multimedia files related to the LSTM music composition project. The files are in MP3 (hi-resolution 128kbps and low resolution 32kbps) and MIDI. A helpful reference document for understanding the compositions is IDSIA Technical Report IDSIA-07-02, A First Look at Music Composition using LSTM Recurrent Neural Networks [postscript or pdf].
music  improvisation  generative-art  hey-I-know-this-guy  nudge-targets  consider:performance-measures
june 2015 by Vaguery
Untangling the roles of parasites in food webs with generative network models | bioRxiv
Food webs represent the set of consumer-resource interactions among a set of species that co-occur in a habitat, but most food web studies have omitted parasites and their interactions. Recent studies have provided conflicting evidence on whether including parasites changes food web structure, with some suggesting that parasitic interactions are structurally distinct from those among free-living species while others claim the opposite. Here, we describe a principled method for understanding food web structure that combines an efficient optimization algorithm from statistical physics called parallel tempering with a probabilistic generalization of the empirically well-supported food web niche model. This generative model approach allows us to rigorously estimate the degree to which interactions that involve parasites are statistically distinguishable from interactions among free-living species, whether parasite niches behave similarly to free-living niches, and the degree to which existing hypotheses about food web structure are naturally recovered. We apply this method to the well-studied Flensburg Fjord food web and show that while predation on parasites, concomitant predation of parasites, and parasitic intraguild trophic interactions are largely indistinguishable from free-living predation interactions, parasite-host interactions are different. These results provide a powerful new tool for evaluating the impact of classes of species and interactions on food web structure to shed new light on the roles of parasites in food webs.
food-webs  ecology  simulation  complexology  hey-I-know-this-guy  nudge-targets  consider:robustness  rather-interesting  inference  statistics  network-theory
june 2015 by Vaguery
[1407.3999] Quantitative patterns in drone wars
Attacks by drones (i.e., unmanned combat air vehicles) continue to generate heated political and ethical debates. Here we examine instead the quantitative nature of drone attacks, focusing on how their intensity and frequency compares to other forms of human conflict. Instead of the power-law distribution found recently for insurgent and terrorist attacks, the severity of attacks is more akin to lognormal and exponential distributions, suggesting that the dynamics underlying drone attacks lie beyond these other forms of human conflict. Meanwhile the pattern in the timing of attacks is consistent with one side having almost complete control. We show that these novel features can be reproduced and understood using a generative mathematical model in which resource allocation to the dominant side is regulated through a feedback loop.
power-laws  hey-I-know-this-guy  war  empirical-economics
june 2015 by Vaguery
Memory with memory
Based in part on observations about the incremental nature of most state changes in biological systems, we introduce the idea of Memory with Memory in Genetic Programming (GP), where we use "soft" assignments to registers instead of the "hard" assignments used in most computer science (including traditional GP). Instead of having the new value completely overwrite the old value of the register, these soft assignments combine the old and new values.

We then report on extensive empirical tests (a total of 12,800 runs) on symbolic regression problems where Memory with Memory GP almost always does as well as traditional GP, while significantly outperforming it in several cases. Memory with Memory GP also tends to be far more consistent, having much less variation in its best-of-run fitnesses than traditional GP. The data suggest that Memory with Memory GP works by successively refining an approximate solution to the target problem. This means it can continue to improve (if slowly) over time, but that it is less likely to get the sort of exact solution that one might find with traditional GP. The use of soft assignment also means that Memory with Memory GP is much less likely to have truly ineffective code, but the action of successive refinement of approximations means that the average program size is often larger than with traditional GP.
from-the-horse's-mouth  in-light-of-West-Eberhard-book  genetic-programming  the-mangle-in-practice  contingency  simulation  rather-interesting  hey-I-know-this-guy  x-2  nudge-targets  consider:do-it-with-Push
june 2015 by Vaguery
Understanding Society: The luminaries and the researcher
So the position I am led to is this. Social research requires theories of how social processes work. It would be foolish to ignore the excellent work of theorizing various aspects of the social world offered by the luminaries. But it would also be foolish to imagine that any one of these theoretical frameworks is total and complete. Rather, the researcher should be eclectic, pluralistic, and curious when it comes to making use of social theory to make sense of a particular range of complex social activity.
worklife  social-norms  sociology  Coscience  disintermediation-targets  hey-I-know-this-guy
november 2014 by Vaguery
[1304.6257] An Evolutionary Algorithm Approach to Link Prediction in Dynamic Social Networks
Many real world, complex phenomena have underlying structures of evolving networks where nodes and links are added and removed over time. A central scientific challenge is the description and explanation of network dynamics, with a key test being the prediction of short and long term changes. For the problem of short-term link prediction, existing methods attempt to determine neighborhood metrics that correlate with the appearance of a link in the next observation period. Recent work has suggested that the incorporation of topological features and node attributes can improve link prediction. We provide an approach to predicting future links by applying Covariance Matrix Adaptation Evolution Strategy (CMA-ES) to optimize weights which are used in a linear combination of sixteen neighborhood and node similarity indices. We examine a large dynamic social network with over 106 nodes (Twitter reciprocal reply networks), both as a test of our general method and as a problem of scientific interest in itself. Our method exhibits fast convergence and high levels of precision for the top twenty predicted links, and to our knowledge, strongly outperforms all extant methods. Based on our findings, we suggest possible factors which may be driving the evolution of Twitter reciprocal reply networks.
social-networks  evolutionary-algorithms  hey-I-know-this-guy  prediction  interesting  nudge-targets  consider:expanded-ontology
december 2013 by Vaguery
[1310.5017] Synthetic biocomputation design using supervised gene regulatory networks
The potential of synthetic biology techniques for designing complex cellular circuits able to solve complicated computations opens a whole domain of exploration, beyond experiments and theory. Such cellular circuits could be used to carry out hard tasks involving decision-making, storage of information, or signal processing. Since Gene Regulatory Networks (GRNs) are the best known technical approach to synthetic designs, it would be desirable to know in advance the potential of such circuits in performing tasks and how classical approximations dealing with neural networks can be translated into GRNs. In this paper such a potential is analyzed. Here we show that feed-forward GRNs are capable of performing classic machine intelligence tasks. Therefore, two important milestones in the success of Artificial Neural Networks are reached for models of GRNs based on Hill equations, namely the back-propagation algorithm and the proof that GRNs can approximate arbitrary positive functions. Potential extensions and implications for synthetic designs are outlined.
engineering-design  genetic-programming-target  systems-biology  biological-engineering  nudge-targets  hey-I-know-this-guy  artificial-life
december 2013 by Vaguery
[1310.8220] Prediction of highly cited papers
In an article written five years ago [arXiv:0809.0522], we described a method for predicting which scientific papers will be highly cited in the future, even if they are currently not highly cited. Applying the method to real citation data we made predictions about papers we believed would end up being well cited. Here we revisit those predictions, five years on, to see how well we did. Among the over 2000 papers in our original data set, we examine the fifty that, by the measures of our previous study, were predicted to do best and we find that they have indeed received substantially more citations in the intervening years than other papers, even after controlling for the number of prior citations. On average these top fifty papers have received 23 times as many citations in the last five years as the average paper in the data set as a whole, and 15 times as many as the average paper in a randomly drawn control group that started out with the same number of citations. Applying our prediction technique to current data, we also make new predictions of papers that we believe will be well cited in the next few years.
december 2013 by Vaguery
[1309.3323] Mapping Mutable Genres in Structurally Complex Volumes
To mine large digital libraries in humanistically meaningful ways, scholars need to divide them by genre. This is a task that classification algorithms are well suited to assist, but they need adjustment to address the specific challenges of this domain. Digital libraries pose two problems of scale not usually found in the article datasets used to test these algorithms. 1) Because libraries span several centuries, the genres being identified may change gradually across the time axis. 2) Because volumes are much longer than articles, they tend to be internally heterogeneous, and the classification task needs to begin with segmentation. We describe a multi-layered solution that trains hidden Markov models to segment volumes, and uses ensembles of overlapping classifiers to address historical change. We test this approach on a collection of 469,200 volumes drawn from HathiTrust Digital Library. To demonstrate the humanistic value of these methods, we extract 32,209 volumes of fiction from the digital library, and trace the changing proportions of first- and third-person narration in the corpus. We note that narrative points of view seem to have strong associations with particular themes and genres.
digital-humanities  classification  natural-language-processing  feature-extraction  nudge-targets  hey-I-know-this-guy
november 2013 by Vaguery
per page:    204080120160

Copy this bookmark:

description:

tags: