representation   4965

« earlier    

TIME International Symposium on Temporal Representation and Reasoning
TIME has been for more than twenty years the only yearly multidisciplinary international event dedicated to the topic of time in computer science. The purpose of the symposium is to bring together active researchers in different research areas involving temporal representation and reasoning. The symposium also welcomes research papers on the related topics of spatial and spatio-temporal representation and reasoning. In the early years, most contributions came from the Artificial Intelligence community, but the number of contributions from other areas such as Temporal Logic and Verification and  partly from Temporal Databases has been increasing in the last years.
time  temporal  symposium  representation  reasoning 
yesterday by areich
Industry coalition forms to banish 'manels' from advertising events by 2021
Diversity and Inclusion at Conferences and Events (Dice) has launched with a 10-point charter laying out how event organisers can promote inclusivity, avoid tokenism and finally banish all-white male panels (or "manels") by 2021.

Based on the 2010 UK Equality Act, the charter looks to help those producing industry events to "prioritise an inclusive approach from the outset". Event managers can submit their agenda to receive an official Dice certification.

The charter covers three elements: line-up, content and marketing. Within this there are guidelines around race, age, gender, class, disability and neurodiversity, in addition to the accessibility of venues. The action plan will be supported by limited free consultations with organisers when necessary.
by:JohnGlenday  from:TheDrum  diversity  representation  advertising  geo:UnitedKingdom 
2 days ago by owenblacker
Fascinated to Presume: In Defense of Fiction | by Zadie Smith | The New York Review of Books
Zadie Smith: "Full disclosure: what insults my soul is the idea—popular in the culture just now, and presented in widely variant degrees of complexity—that we can and should write only about people who are fundamentally “like” us: racially, sexually, genetically, nationally, politically, personally. That only an intimate authorial autobiographical connection with a character can be the rightful basis of a fiction. I do not believe that. I could not have written a single one of my books if I did. But I feel no sense of triumph in my apostasy. It might well be that we simply don’t want or need novels like mine anymore, or any of the kinds of fictions that, in order to exist, must fundamentally disagree with the new theory of “likeness.” It may be that the whole category of what we used to call fiction is becoming lost to us. And if enough people turn from the concept of fiction as it was once understood, then fighting this transformation will be like going to war against the neologism “impactful” or mourning the loss of the modal verb “shall.” As it is with language, so it goes with culture: what is not used or wanted dies. What is needed blooms and spreads."
writing  representation  publishing  identity  appropriation 
2 days ago by jbushnell
Weregeek » 2020 » February » 18
On dark-skinned = villainy in TTRPGs.
500  representation 
3 days ago by dsalo
[1812.08434] Graph Neural Networks: A Review of Methods and Applications
Lots of learning tasks require dealing with graph data which contains rich relation information among elements. Modeling physics system, learning molecular fingerprints, predicting protein interface, and classifying diseases require a model to learn from graph inputs. In other domains such as learning from non-structural data like texts and images, reasoning on extracted structures, like the dependency tree of sentences and the scene graph of images, is an important research topic which also needs graph reasoning models. Graph neural networks (GNNs) are connectionist models that capture the dependence of graphs via message passing between the nodes of graphs. Unlike standard neural networks, graph neural networks retain a state that can represent information from its neighborhood with arbitrary depth. Although the primitive GNNs have been found difficult to train for a fixed point, recent advances in network architectures, optimization techniques, and parallel computation have enabled successful learning with them. In recent years, systems based on variants of graph neural networks such as graph convolutional network (GCN), graph attention network (GAT), gated graph neural network (GGNN) have demonstrated ground-breaking performance on many tasks mentioned above. In this survey, we provide a detailed review over existing graph neural network models, systematically categorize the applications, and propose four open problems for future research.
machine-learning  representation  review  rather-interesting  to-write-about  to-simulate  consider:genetic-programming  consider:functional-transformations 
6 days ago by Vaguery
[1809.02836] Context-Free Transductions with Neural Stacks
This paper analyzes the behavior of stack-augmented recurrent neural network (RNN) models. Due to the architectural similarity between stack RNNs and pushdown transducers, we train stack RNN models on a number of tasks, including string reversal, context-free language modelling, and cumulative XOR evaluation. Examining the behavior of our networks, we show that stack-augmented RNNs can discover intuitive stack-based strategies for solving our tasks. However, stack RNNs are more difficult to train than classical architectures such as LSTMs. Rather than employ stack-based strategies, more complex networks often find approximate solutions by using the stack as unstructured memory.
neural-networks  artificial-life  representation  time-series  automata  to-write-about  to-simulate  unconventional-computing 
6 days ago by Vaguery
[1906.01615] Sequential Neural Networks as Automata
This work attempts to explain the types of computation that neural networks can perform by relating them to automata. We first define what it means for a real-time network with bounded precision to accept a language. A measure of network memory follows from this definition. We then characterize the classes of languages acceptable by various recurrent networks, attention, and convolutional networks. We find that LSTMs function like counter machines and relate convolutional networks to the subregular hierarchy. Overall, this work attempts to increase our understanding and ability to interpret neural networks through the lens of theory. These theoretical insights help explain neural computation, as well as the relationship between neural networks and natural language grammar.
neural-networks  rather-interesting  update-dynamics  representation  via:cshalizi  to-read  to-write-about  to-simulate  automata  unconventional-computation  but-is-it-really? 
6 days ago by Vaguery
Not Quite Banned: Soft Censorship That Makes LGBTQIA+ Stories Disappear | School Library Journal
“Librarians are fighting very real fights in our districts for funding, for resources, sometimes against [people] hostile to your existence at all in school. So I get that,” London continued. “Some librarians have to choose.…There are lots of great authors who will not cause any controversy. It’s an easy decision for a lot of people. It’s a deadly decision for a lot of students.”
LGBTQ  libraries  first-amendment  politics  representation  republiklan_agenda 
16 days ago by kadymae
[1809.08898] Rectilinear Shortest Paths Among Transient Obstacles
This paper presents an optimal Θ(nlogn) algorithm for determining time-minimal rectilinear paths among n transient rectilinear obstacles. An obstacle is transient if it exists in the scene only for a specific time interval, i.e., it appears and then disappears at specific times. Given a point robot moving with bounded speed among transient rectilinear obstacles and a pair of points s, d, we determine a time-minimal, obstacle-avoiding path from s to d. The main challenge in solving this problem arises as the robot may be required to wait for an obstacle to disappear, before it can continue moving toward the destination. Our algorithm builds on the continuous Dijkstra paradigm, which simulates propagating a wavefront from the source point. We also solve a query version of this problem. For this, we build a planar subdivision with respect to a fixed source point, so that minimum arrival time to any query point can be reported in O(logn) time, using point location for the query point in this subdivision.
computational-geometry  computational-complexity  planning  rather-interesting  to-understand  representation  wavelets  to-simulate  consider:representation 
17 days ago by Vaguery

« earlier    

related tags

(sorry)  2019  351  360  500  601  911  ablegamers  acting  adjacent-possible  advertising  agda  agency  ai  albert  algorithms  amazon  americandirt  animation  anti-racism  antiracism  appropriation  arabs  architecture  art  artificial-life  asian-american  asianamerica  automata  bbc  bert  bigdata  binary  biopics  blackness  blackwomen  blog  books  boolean-networks  boolean  british  brownfacing  bureaucracy  but-is-it-really?  by:almazohene  by:charlesyu  by:johnglenday  by:junesarpong  canada  cause-and-effect  census  character  characterisation  chatbot  cinema  citizenship  climate_change  combinatorics  comics  communication  compilers  computational-complexity  computational-geometry  computer-science  conceptualization  concurrency  consider:constants  consider:feature-discovery  consider:functional-transformations  consider:genetic-programming  consider:irrational-values  consider:operations  consider:performance-measures  consider:representation  consider:sense-of-scale  consider:simulation  consider:subsetting  consider:truchet-tiling  constraint-satisfaction  course.use  creativity  creatures  culture  data  datatypes  davidschmidt  death  deep-learning  democracy  design  development  digital_media  dimer-models  disabled  disney  diversity  domain-specific  domino-tiling  effect  empathy  engineering-criticism  exploding-dots  exploration  feature-construction  file  film  filmmaking  first-amendment  formal-languages  format  fraud  from:campaign  from:huffpost  from:thedrum  gamer  games  gender  generative  geo:unitedkingdom  geog107  gerrymandering  google  googleblog  graph  graphics  gretagerwig  grin  group-theory  hashing  haskell  headcanon  heavenly  hierarchy  history-of-engineering  history  hpc  hypergraphs  ic  identity  ideology  idris  illustration  image-processing  indirect-representations  indirection  individuality  inequality  information-theory  information  intermediate  issue_salience  jackson  jeaninecummins  journalism  json  knot-theory  language  latent  latinx  lgbt  lgbtq  libraries  ling  literature  litofsport  littlewomen  looking-to-see  machine-learning  machinelearning  manifold  martinscorsese  marvel  mass  mathematical-recreations  matrices  mcu  media  mexico  ml  mobility  model  multivalued-logics  muslim  myriamgurba  neural-networks  nlp  notation  number-theory  object  ontology  opendata  opinionpiece  other  othering  package  pedagogy  peter  philosophy-of-engineering  plane-geometry  planning  poc  political_parties  politics  polyamory  probabilistic-languages  probability-theory  problems  publishing  quantum-computing  quantum-physics  quantums  queerness  race  racegender  racism  radix  random-sampling  rather-interesting  rather-odd  rdf  reaction-networks  readerownership  reading  readingin  reasoning  recruitment  relationships  republiklan_agenda  resolution  responsiveness  review  reviews  rhetoric  romance-novels  scale  scientific  searchengines  self  selfsupervised  serialization  sex_work  socialjustice  software  softwaredevelopment  sports  standard  statistical-mechanics  stereotypes  syllabus  symbolic-forms  sympathy  symposium  teaching  team  television  temporal  thane  time-series  time  to-read  to-simulate  to-understand  to-write-about  training  transgender  translation  trope-crossdressing  tropes-crossdressing  truestories  tv  tv:modernlove  uk  unconventional-computation  unconventional-computing  universality  update-dynamics  user-experience  video  visual  walking  wavelets  web  welfare  whiteness  women's-sports  work  writing  yafiction 

Copy this bookmark: