Bookmarks

2015-05-26

場所参照表現タグ付きコーパス Ver 0.1 (2015/05/25)

http://www.cl.ecei.tohoku.ac.jp/~matsuda/LRE_corpus/

概要

本コーパスは，Twitterからランダムにサンプリングしたテキストに現れる，「特定の場所を著者が想定している」と判断できる表現に対して，実際にどのエンティティを指しているかを人手で判断しエンティティ情報を付与したコーパスです．GeoNLPなどのジオパーズシステム，エンティティリンキングシステム等の開発や評価に利用することを想定して構築されました．

Computing numeric representations of words in a high-dimensional space - Google Inc.

http://www.freepatentsonline.com/9037464.html

United States Patent 9037464

Inventors:
Mikolov, Tomas (Jersey City, NJ, US)
Chen, Kai (San Bruno, CA, US)
Corrado, Gregory S. (San Francisco, CA, US)
Dean, Jeffrey A. (Palo Alto, CA, US)

word2vec 関連の特許。

ConvNetJS MNIST demo

http://cs.stanford.edu/people/karpathy/convnetjs/demo/mnist.html

This demo trains a Convolutional Neural Network on the MNIST digits dataset in your browser, with nothing but Javascript.

手書き数字の画像データを JavaScript のプログラムで学習するデモ。
学習中の各層（conv, pool, softmax）の重みやアクティベーション値をリアルタイムで見ることができる。

The Unreasonable Effectiveness of Recurrent Neural Networks

http://karpathy.github.io/2015/05/21/rnn-effectiveness/

We'll train RNNs to generate text character by character and ponder the question "how is that even possible?"

Andrej Karpathy Academic Website

http://cs.stanford.edu/people/karpathy/

Andrej Karpathy
Stanford Computer Science Ph.D. student

karpathy/convnetjs · GitHub

https://github.com/karpathy/convnetjs

ConvNetJS is a Javascript implementation of Neural networks, together with nice browser-based demos.

ConvNetJS: Deep Learning in your browser

http://cs.stanford.edu/people/karpathy/convnetjs/

ConvNetJS is a Javascript library for training Deep Learning models (mainly Neural Networks) entirely in your browser. Open a tab and you're training. No software requirements, no compilers, no installations, no GPUs, no sweat.

2015-05-16

分散共分散行列と相関行列

https://upo-net.ouj.ac.jp/tokei/xml/k3_06022.xml

2015-05-01

強化学習をベイズで理解する

https://www.sat.t.u-tokyo.ac.jp/~mak/20140319-makino.pdf

2015-04-27

GloVe: Global Vectors for Word Representation

http://nlp.stanford.edu/projects/glove/

GloVe is an unsupervised learning algorithm for obtaining vector representations for words. Training is performed on aggregated global word-word co-occurrence statistics from a corpus, and the resulting representations showcase interesting linear substructures of the word vector space.

2015-04-20

Echo state network - Wikipedia, the free encyclopedia

http://en.wikipedia.org/wiki/Echo_state_network

The echo state network (ESN) is a recurrent neural network with a sparsely connected hidden layer (with typically 1% connectivity). The connectivity and weights of hidden neurons are randomly assigned and are fixed. The weights of output neurons can be learned so that the network can (re)produce specific temporal patterns.

Reservoir computing - Wikipedia, the free encyclopedia

http://en.wikipedia.org/wiki/Reservoir_computing

Reservoir computing is a framework for computation like a neural network.

The “echo state” approach to analysing and training recurrent neural networks – with an Erratum note

http://web.info.uvt.ro/~dzaharie/cne2013/proiecte/tehnici/ReservoirComputing/EchoStatesTechRep.pdf

This is a corrected version of the technical report H. Jaeger(2001): The ”echo state” approach to analysing and training recurrent neural networks. GMD Report 148, German National Research Center for Information Techno logy, 2001.

An overview of reservoir computing: theory, applications and implementations

https://www.elen.ucl.ac.be/Proceedings/esann/esannpdf/es2007-8.pdf

This tutorial will give an overview of current research on theory, application and implementations of Reservoir Computing.

Frontiers | MACOP modular architecture with control primitives | Frontiers in Computational Neuroscience

http://journal.frontiersin.org/article/10.3389/fncom.2013.00099/full

2.4.2. Echo state networks
We use an Echo State Network (ESN) (Jaeger, 2001) as inverse model. An ESN is composed of a discrete-time recurrent neural network [commonly called the reservoir because ESNs belong to the class of Reservoir Computing techniques (Schrauwen et al., 2007)] and a linear readout layer which maps the state of the reservoir to the desired output.