papers   24326

« earlier    

[1701.02827] Strong Functional Representation Lemma and Applications to Coding Theorems
This paper shows that for any random variables $X$ and $Y$, it is possible to represent $Y$ as a function of $(X,Z)$ such that $Z$ is independent of $X$ and $I(X;Z|Y)\le\log(I(X;Y)+1)+4$ bits. We use this strong functional representation lemma (SFRL) to establish a bound on the rate needed for one-shot exact channel simulation for general (discrete or continuous) random variables, strengthening the results by Harsha et al. and Braverman and Garg, and to establish new and simple achievability results for one-shot variable-length lossy source coding, multiple description coding and Gray-Wyner system. We also show that the SFRL can be used to reduce the channel with state noncausally known at the encoder to a point-to-point channel, which provides a simple achievability proof of the Gelfand-Pinsker theorem.
papers  to-read  information-theory  simulation 
13 hours ago by mraginsky
[1611.01116] Binary Paragraph Vectors
Recently Le & Mikolov described two log-linear models, called Paragraph Vector, that can be used to learn state-of-the-art distributed representations of documents. Inspired by this work, we present Binary Paragraph Vector models: simple neural networks that learn short binary codes for fast information retrieval. We show that binary paragraph vectors outperform autoencoder-based binary codes, despite using fewer bits. We also evaluate their precision in transfer learning settings, where binary codes are inferred for documents unrelated to the training corpus. Results from these experiments indicate that binary paragraph vectors can capture semantics relevant for various domain-specific documents. Finally, we present a model that simultaneously learns short binary codes and longer, real-valued representations. This model can be used to rapidly retrieve a short list of highly relevant documents from a large document collection.
IR  embeddings  index  papers 
15 hours ago by foodbaby
[1806.05722] Non-asymptotic Identification of LTI Systems from a Single Trajectory
We consider the problem of learning a realization for a linear time-invariant (LTI) dynamical system from input/output data. Given a single input/output trajectory, we provide finite time analysis for learning the system's Markov parameters, from which a balanced realization is obtained using the classical Ho-Kalman algorithm. By proving a stability result for the Ho-Kalman algorithm and combining it with the sample complexity results for Markov parameters, we show how much data is needed to learn a balanced realization of the system up to a desired accuracy with high probability.
papers  to-read  system-identification  control-theory 
yesterday by mraginsky

« earlier    

related tags

academic  academic_paper  academics  acolyer  aging  ai  algorithms  animation  archives  authentication  author:c._brian_bucklew  author:jason_grinblat  bayes  beyondcorp  book  bookends  bookmarks_bar  books  buddhism  build  calibration  caves.of.qud  certificatemanagement  chart  charts  citation-graph  cnn  compiler  compilers  compsci  compute  computer_science  computer_vision  conferences  control-theory  crypto  cs  cv  daily  data-processing  data-science  data  data_structures  data_tool  database  databases  datascience  deep-learning-theory  deep-learning  deepleab  deeplearning  design  detection  dev  diet  digitising  discussion  dl  downloading  dunlop  edu  ellsberg  emacs  embeddings  engineering  fatml  feed  finance  fp  fractals  fraud  generation  google  graphic  graphics  graphs  hacks  haskell  health  hs  image  index  information-theory  infosec  investigations  ir  issue  journalism  journals  jupyter  kaggle  landmark  language_design  learn  learning  list  lists  logging  machine-learning  machine_learning  machinelearning  make  math  mathematics  medicine  mit  ml  neural-networks  neural_networks  nih  nlp  notation  notebook  number_theory  ontology  open.access  orgmode  panama  paper  paradise  parsing  pattern  pdf  pentagon  philosophy  pinterest  pl  placecal  plot  presentations  procedural  proceedings  program-analysis  programming-languages  programming  programming_languages  psychedelics  psychopharmacology  quantized  ranking  reading  readinglist  recommendation  ref  reference  reference_management  regex  reinforcement-learning  reinforcement  rendering  reproducibility  research  resource  resources  retractions  reverse  rubber  saleft  science  scihub  scripting  security  simulation  slowmotion  special  sql  staticanalysis  stochastic-control  system-identification  technology  tf-idf  to-read  todo  tool  tools  toread  tprc  trump  tutorial  twitter  types  unabomber  university  volumes  web  webapp  windows  word2vec  work  wrapping  writing 

Copy this bookmark:



description:


tags: