Modeling Documents with a Deep Boltzmann Machine

Size: px
Start display at page:

Download "Modeling Documents with a Deep Boltzmann Machine"

Transcription

1 Modeling Documents with a Deep Boltzmann Machine Nitish Srivastava, Ruslan Salakhutdinov & Geoffrey Hinton UAI 2013 Presented by Zhe Gan, Duke University November 14, / 15

2 Outline Replicated Softmax Model (Hinton and Salakhutdinov, 2009). Undirected topic model Over-Replicated Softmax Model (Srivastava et al., 2013) Only has one more parameter than the Replicated Softmax Related work DocNADE (Larochelle and Lauly, 2012) Experiments 20 Newsgroup & Reuters Corpus Volume I 2 / 15

3 Replicated Softmax Model Let K be the vocabulary size, N be the number of words, F be the number of hidden topic features A document can be represented as a binary matrix V R N K, with v ik = 1 if visible unit i takes the k th value. Hidden units are h R F. Energy function defined as E(V, h; θ) = N F i=1 =1 k=1 K W ik h v ik N i=1 k=1 K v ik b ik N Key assumption: Softmax units share the same set of weights. E(V, h; θ) = F =1 k=1 K W k h ˆv k K ˆv k b k N k=1 where ˆv k = N i=1 v ik, denotes the count for the k th word. F h a =1 F h a (1) =1 3 / 15

4 Replicated Softmax Model Conditional distribution: ( K ) p(h = 1 V) =σ W k ˆv k + Na (2) k=1 ( F ) exp =1 W kh + b k p(v ik = 1 h) = ( K k =1 exp F (3) =1 W k h + b k ) Figure : The Replicated Softmax model. 4 / 15

5 Over-Replicated Softmax Model Now, add the second hidden layer, which consists of M softmax units, i.e. H (2) R M K. Tie the 1-st and 2-nd layer weights to be the same. No additional parameters. In essence, the 2-nd layer can be considered as missing / pseudo words. Figure : The Over-Replicated Softmax model. 5 / 15

6 Over-Replicated Softmax Model Now, the energy is defined as E(V, h (1), H (2) ) = where ĥ(2) k = M pseudo document. i=1 h(2) ik F K =1 k=1 K ( k=1 ( W k h (1) ˆv k + ĥ(2) k ˆv k + ĥ(2) k ) ) b k (M + N) F =1 h (1) a is the pseudo count and M is the length of the Given L documents, the gradient of the log-likelihood is 1 L L l=1 log P(V) [ = E Pdata (ˆv k + W ĥ(2) k k )h(1) ] E Pmodel [ ] (ˆv k + ĥ(2) k )h(1) 6 / 15

7 Learning and Inference Since P data (h, V) = P(h V; θ)p data (V), we use mean-field inference to estimate the data-dependent expectations. Q MF (h V; µ) = F =1 q(h (1) V) M i=1 q(h (2) i V) (4) To be specific, q(h (1) = 1) = µ (1) and q(h (2) ik = 1) = µ (2) k,then ( µ (1) K ( ) ) =σ W k ˆv k + Mµ (2) (5) µ (2) k = k=1 ( F ) exp =1 W kµ (1) ( K k =1 exp F =1 W k µ(1) k ) (6) 7 / 15

8 Learning and Inference Use an MCMC based stochastic approximation procedure to approximate the model expectation, i.e. Contrastive Divergence. A typical update: V t {h (1) t, H (2) t } V t+1 {h (1) t+1, H(2) t+1 }. Use a point estimate at t + 1 to approximate the model distribution. Pre-training: Train an RBM with scaled weights. ( = 1 V) = σ (1 + M K ) N ) W k ˆv k P(h (1) Learning: already given in Equation (5) k=1 Choosing M: Needs to be chosen using a validation set. (7) 8 / 15

9 Related Work NADE (Neural Autoregressive Distribution Estimation): a generative model over binary vector v {0, 1} D. p(v i = 1 v <i ) = σ (b i + V i,: h i ) (8) h i (v i ) = σ(c + W :,<i v <i ) (9) where W R H D and V R D H. Modeling each conditional using a neural network. 9 / 15

10 Related Work Document NADE p(v i = w v <i ) = exp(b w + V w,: h i ) w exp(b w + V w,:h i ) (10) h i (v i ) = σ(c + k<i W :,vk ) (11) The large softmax over words is replaced by a probabilistic tree in which each path from the root to a leaf corresponds to a word. 10 / 15

11 Experiments Mini-batch size: 128; M = 100; Topic Number: 128 The model implemented on GPUs; pretraining took 3-4 hours, the proper training took hours. 11 / 15

12 Experiments - Document Retrieval 12 / 15

13 Experiments The Over-Replicated Softmax model works well on short documents. 13 / 15

14 Experiments 14 / 15

15 References I G. E. Hinton and R. Salakhutdinov. Replicated softmax: an undirected topic model. NIPS, H. Larochelle and S. Lauly. A neural autoregressive topic model. NIPS, N. Srivastava, R. R. Salakhutdinov, and G. E. Hinton. Modeling documents with deep boltzmann machines. UAI, / 15

Fast Inference and Learning for Modeling Documents with a Deep Boltzmann Machine

Fast Inference and Learning for Modeling Documents with a Deep Boltzmann Machine Fast Inference and Learning for Modeling Documents with a Deep Boltzmann Machine Nitish Srivastava nitish@cs.toronto.edu Ruslan Salahutdinov rsalahu@cs.toronto.edu Geoffrey Hinton hinton@cs.toronto.edu

More information

Deep Boltzmann Machines

Deep Boltzmann Machines Deep Boltzmann Machines Ruslan Salakutdinov and Geoffrey E. Hinton Amish Goel University of Illinois Urbana Champaign agoel10@illinois.edu December 2, 2016 Ruslan Salakutdinov and Geoffrey E. Hinton Amish

More information

Replicated Softmax: an Undirected Topic Model. Stephen Turner

Replicated Softmax: an Undirected Topic Model. Stephen Turner Replicated Softmax: an Undirected Topic Model Stephen Turner 1. Introduction 2. Replicated Softmax: A Generative Model of Word Counts 3. Evaluating Replicated Softmax as a Generative Model 4. Experimental

More information

A graph contains a set of nodes (vertices) connected by links (edges or arcs)

A graph contains a set of nodes (vertices) connected by links (edges or arcs) BOLTZMANN MACHINES Generative Models Graphical Models A graph contains a set of nodes (vertices) connected by links (edges or arcs) In a probabilistic graphical model, each node represents a random variable,

More information

Greedy Layer-Wise Training of Deep Networks

Greedy Layer-Wise Training of Deep Networks Greedy Layer-Wise Training of Deep Networks Yoshua Bengio, Pascal Lamblin, Dan Popovici, Hugo Larochelle NIPS 2007 Presented by Ahmed Hefny Story so far Deep neural nets are more expressive: Can learn

More information

Au-delà de la Machine de Boltzmann Restreinte. Hugo Larochelle University of Toronto

Au-delà de la Machine de Boltzmann Restreinte. Hugo Larochelle University of Toronto Au-delà de la Machine de Boltzmann Restreinte Hugo Larochelle University of Toronto Introduction Restricted Boltzmann Machines (RBMs) are useful feature extractors They are mostly used to initialize deep

More information

An Efficient Learning Procedure for Deep Boltzmann Machines

An Efficient Learning Procedure for Deep Boltzmann Machines ARTICLE Communicated by Yoshua Bengio An Efficient Learning Procedure for Deep Boltzmann Machines Ruslan Salakhutdinov rsalakhu@utstat.toronto.edu Department of Statistics, University of Toronto, Toronto,

More information

The Origin of Deep Learning. Lili Mou Jan, 2015

The Origin of Deep Learning. Lili Mou Jan, 2015 The Origin of Deep Learning Lili Mou Jan, 2015 Acknowledgment Most of the materials come from G. E. Hinton s online course. Outline Introduction Preliminary Boltzmann Machines and RBMs Deep Belief Nets

More information

Lecture 16 Deep Neural Generative Models

Lecture 16 Deep Neural Generative Models Lecture 16 Deep Neural Generative Models CMSC 35246: Deep Learning Shubhendu Trivedi & Risi Kondor University of Chicago May 22, 2017 Approach so far: We have considered simple models and then constructed

More information

Restricted Boltzmann Machines for Collaborative Filtering

Restricted Boltzmann Machines for Collaborative Filtering Restricted Boltzmann Machines for Collaborative Filtering Authors: Ruslan Salakhutdinov Andriy Mnih Geoffrey Hinton Benjamin Schwehn Presentation by: Ioan Stanculescu 1 Overview The Netflix prize problem

More information

Chapter 20. Deep Generative Models

Chapter 20. Deep Generative Models Peng et al.: Deep Learning and Practice 1 Chapter 20 Deep Generative Models Peng et al.: Deep Learning and Practice 2 Generative Models Models that are able to Provide an estimate of the probability distribution

More information

Restricted Boltzmann Machines

Restricted Boltzmann Machines Restricted Boltzmann Machines Boltzmann Machine(BM) A Boltzmann machine extends a stochastic Hopfield network to include hidden units. It has binary (0 or 1) visible vector unit x and hidden (latent) vector

More information

Stochastic Gradient Estimate Variance in Contrastive Divergence and Persistent Contrastive Divergence

Stochastic Gradient Estimate Variance in Contrastive Divergence and Persistent Contrastive Divergence ESANN 0 proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. Bruges (Belgium), 7-9 April 0, idoc.com publ., ISBN 97-7707-. Stochastic Gradient

More information

Density estimation. Computing, and avoiding, partition functions. Iain Murray

Density estimation. Computing, and avoiding, partition functions. Iain Murray Density estimation Computing, and avoiding, partition functions Roadmap: Motivation: density estimation Understanding annealing/tempering NADE Iain Murray School of Informatics, University of Edinburgh

More information

An Efficient Learning Procedure for Deep Boltzmann Machines Ruslan Salakhutdinov and Geoffrey Hinton

An Efficient Learning Procedure for Deep Boltzmann Machines Ruslan Salakhutdinov and Geoffrey Hinton Computer Science and Artificial Intelligence Laboratory Technical Report MIT-CSAIL-TR-2010-037 August 4, 2010 An Efficient Learning Procedure for Deep Boltzmann Machines Ruslan Salakhutdinov and Geoffrey

More information

COMP9444 Neural Networks and Deep Learning 11. Boltzmann Machines. COMP9444 c Alan Blair, 2017

COMP9444 Neural Networks and Deep Learning 11. Boltzmann Machines. COMP9444 c Alan Blair, 2017 COMP9444 Neural Networks and Deep Learning 11. Boltzmann Machines COMP9444 17s2 Boltzmann Machines 1 Outline Content Addressable Memory Hopfield Network Generative Models Boltzmann Machine Restricted Boltzmann

More information

Introduction to Restricted Boltzmann Machines

Introduction to Restricted Boltzmann Machines Introduction to Restricted Boltzmann Machines Ilija Bogunovic and Edo Collins EPFL {ilija.bogunovic,edo.collins}@epfl.ch October 13, 2014 Introduction Ingredients: 1. Probabilistic graphical models (undirected,

More information

Learning to Disentangle Factors of Variation with Manifold Learning

Learning to Disentangle Factors of Variation with Manifold Learning Learning to Disentangle Factors of Variation with Manifold Learning Scott Reed Kihyuk Sohn Yuting Zhang Honglak Lee University of Michigan, Department of Electrical Engineering and Computer Science 08

More information

UNSUPERVISED LEARNING

UNSUPERVISED LEARNING UNSUPERVISED LEARNING Topics Layer-wise (unsupervised) pre-training Restricted Boltzmann Machines Auto-encoders LAYER-WISE (UNSUPERVISED) PRE-TRAINING Breakthrough in 2006 Layer-wise (unsupervised) pre-training

More information

Reading Group on Deep Learning Session 4 Unsupervised Neural Networks

Reading Group on Deep Learning Session 4 Unsupervised Neural Networks Reading Group on Deep Learning Session 4 Unsupervised Neural Networks Jakob Verbeek & Daan Wynen 206-09-22 Jakob Verbeek & Daan Wynen Unsupervised Neural Networks Outline Autoencoders Restricted) Boltzmann

More information

Restricted Boltzmann Machines

Restricted Boltzmann Machines Restricted Boltzmann Machines http://deeplearning4.org/rbm-mnist-tutorial.html Slides from Hugo Larochelle, Geoffrey Hinton, and Yoshua Bengio CSC321: Intro to Machine Learning and Neural Networks, Winter

More information

Ruslan Salakhutdinov Joint work with Geoff Hinton. University of Toronto, Machine Learning Group

Ruslan Salakhutdinov Joint work with Geoff Hinton. University of Toronto, Machine Learning Group NON-LINEAR DIMENSIONALITY REDUCTION USING NEURAL NETORKS Ruslan Salakhutdinov Joint work with Geoff Hinton University of Toronto, Machine Learning Group Overview Document Retrieval Present layer-by-layer

More information

Introduction to Deep Neural Networks

Introduction to Deep Neural Networks Introduction to Deep Neural Networks Presenter: Chunyuan Li Pattern Classification and Recognition (ECE 681.01) Duke University April, 2016 Outline 1 Background and Preliminaries Why DNNs? Model: Logistic

More information

Large-Scale Feature Learning with Spike-and-Slab Sparse Coding

Large-Scale Feature Learning with Spike-and-Slab Sparse Coding Large-Scale Feature Learning with Spike-and-Slab Sparse Coding Ian J. Goodfellow, Aaron Courville, Yoshua Bengio ICML 2012 Presented by Xin Yuan January 17, 2013 1 Outline Contributions Spike-and-Slab

More information

Learning Deep Boltzmann Machines using Adaptive MCMC

Learning Deep Boltzmann Machines using Adaptive MCMC Ruslan Salakhutdinov Brain and Cognitive Sciences and CSAIL, MIT 77 Massachusetts Avenue, Cambridge, MA 02139 rsalakhu@mit.edu Abstract When modeling high-dimensional richly structured data, it is often

More information

Gaussian Cardinality Restricted Boltzmann Machines

Gaussian Cardinality Restricted Boltzmann Machines Proceedings of the Twenty-Ninth AAAI Conference on Artificial Intelligence Gaussian Cardinality Restricted Boltzmann Machines Cheng Wan, Xiaoming Jin, Guiguang Ding and Dou Shen School of Software, Tsinghua

More information

arxiv: v1 [cs.ne] 6 May 2014

arxiv: v1 [cs.ne] 6 May 2014 Training Restricted Boltzmann Machine by Perturbation arxiv:1405.1436v1 [cs.ne] 6 May 2014 Siamak Ravanbakhsh, Russell Greiner Department of Computing Science University of Alberta {mravanba,rgreiner@ualberta.ca}

More information

Using Deep Belief Nets to Learn Covariance Kernels for Gaussian Processes

Using Deep Belief Nets to Learn Covariance Kernels for Gaussian Processes Using Deep Belief Nets to Learn Covariance Kernels for Gaussian Processes Ruslan Salakhutdinov and Geoffrey Hinton Department of Computer Science, University of Toronto 6 King s College Rd, M5S 3G4, Canada

More information

Learning Deep Architectures for AI. Part II - Vijay Chakilam

Learning Deep Architectures for AI. Part II - Vijay Chakilam Learning Deep Architectures for AI - Yoshua Bengio Part II - Vijay Chakilam Limitations of Perceptron x1 W, b 0,1 1,1 y x2 weight plane output =1 output =0 There is no value for W and b such that the model

More information

arxiv: v4 [cs.lg] 16 Apr 2015

arxiv: v4 [cs.lg] 16 Apr 2015 REWEIGHTED WAKE-SLEEP Jörg Bornschein and Yoshua Bengio Department of Computer Science and Operations Research University of Montreal Montreal, Quebec, Canada ABSTRACT arxiv:1406.2751v4 [cs.lg] 16 Apr

More information

Unsupervised Learning

Unsupervised Learning CS 3750 Advanced Machine Learning hkc6@pitt.edu Unsupervised Learning Data: Just data, no labels Goal: Learn some underlying hidden structure of the data P(, ) P( ) Principle Component Analysis (Dimensionality

More information

Deep unsupervised learning

Deep unsupervised learning Deep unsupervised learning Advanced data-mining Yongdai Kim Department of Statistics, Seoul National University, South Korea Unsupervised learning In machine learning, there are 3 kinds of learning paradigm.

More information

Maxout Networks. Hien Quoc Dang

Maxout Networks. Hien Quoc Dang Maxout Networks Hien Quoc Dang Outline Introduction Maxout Networks Description A Universal Approximator & Proof Experiments with Maxout Why does Maxout work? Conclusion 10/12/13 Hien Quoc Dang Machine

More information

Bias-Variance Trade-Off in Hierarchical Probabilistic Models Using Higher-Order Feature Interactions

Bias-Variance Trade-Off in Hierarchical Probabilistic Models Using Higher-Order Feature Interactions - Trade-Off in Hierarchical Probabilistic Models Using Higher-Order Feature Interactions Simon Luo The University of Sydney Data61, CSIRO simon.luo@data61.csiro.au Mahito Sugiyama National Institute of

More information

Chapter 16. Structured Probabilistic Models for Deep Learning

Chapter 16. Structured Probabilistic Models for Deep Learning Peng et al.: Deep Learning and Practice 1 Chapter 16 Structured Probabilistic Models for Deep Learning Peng et al.: Deep Learning and Practice 2 Structured Probabilistic Models way of using graphs to describe

More information

Speaker Representation and Verification Part II. by Vasileios Vasilakakis

Speaker Representation and Verification Part II. by Vasileios Vasilakakis Speaker Representation and Verification Part II by Vasileios Vasilakakis Outline -Approaches of Neural Networks in Speaker/Speech Recognition -Feed-Forward Neural Networks -Training with Back-propagation

More information

An efficient way to learn deep generative models

An efficient way to learn deep generative models An efficient way to learn deep generative models Geoffrey Hinton Canadian Institute for Advanced Research & Department of Computer Science University of Toronto Joint work with: Ruslan Salakhutdinov, Yee-Whye

More information

Deep Learning Srihari. Deep Belief Nets. Sargur N. Srihari

Deep Learning Srihari. Deep Belief Nets. Sargur N. Srihari Deep Belief Nets Sargur N. Srihari srihari@cedar.buffalo.edu Topics 1. Boltzmann machines 2. Restricted Boltzmann machines 3. Deep Belief Networks 4. Deep Boltzmann machines 5. Boltzmann machines for continuous

More information

Neural Networks for Machine Learning. Lecture 11a Hopfield Nets

Neural Networks for Machine Learning. Lecture 11a Hopfield Nets Neural Networks for Machine Learning Lecture 11a Hopfield Nets Geoffrey Hinton Nitish Srivastava, Kevin Swersky Tijmen Tieleman Abdel-rahman Mohamed Hopfield Nets A Hopfield net is composed of binary threshold

More information

Deep Generative Models. (Unsupervised Learning)

Deep Generative Models. (Unsupervised Learning) Deep Generative Models (Unsupervised Learning) CEng 783 Deep Learning Fall 2017 Emre Akbaş Reminders Next week: project progress demos in class Describe your problem/goal What you have done so far What

More information

Deep Learning Basics Lecture 8: Autoencoder & DBM. Princeton University COS 495 Instructor: Yingyu Liang

Deep Learning Basics Lecture 8: Autoencoder & DBM. Princeton University COS 495 Instructor: Yingyu Liang Deep Learning Basics Lecture 8: Autoencoder & DBM Princeton University COS 495 Instructor: Yingyu Liang Autoencoder Autoencoder Neural networks trained to attempt to copy its input to its output Contain

More information

Autoencoders and Score Matching. Based Models. Kevin Swersky Marc Aurelio Ranzato David Buchman Benjamin M. Marlin Nando de Freitas

Autoencoders and Score Matching. Based Models. Kevin Swersky Marc Aurelio Ranzato David Buchman Benjamin M. Marlin Nando de Freitas On for Energy Based Models Kevin Swersky Marc Aurelio Ranzato David Buchman Benjamin M. Marlin Nando de Freitas Toronto Machine Learning Group Meeting, 2011 Motivation Models Learning Goal: Unsupervised

More information

The Recurrent Temporal Restricted Boltzmann Machine

The Recurrent Temporal Restricted Boltzmann Machine The Recurrent Temporal Restricted Boltzmann Machine Ilya Sutskever, Geoffrey Hinton, and Graham Taylor University of Toronto {ilya, hinton, gwtaylor}@cs.utoronto.ca Abstract The Temporal Restricted Boltzmann

More information

An Empirical Investigation of Minimum Probability Flow Learning Under Different Connectivity Patterns

An Empirical Investigation of Minimum Probability Flow Learning Under Different Connectivity Patterns An Empirical Investigation of Minimum Probability Flow Learning Under Different Connectivity Patterns Daniel Jiwoong Im, Ethan Buchman, and Graham W. Taylor School of Engineering University of Guelph Guelph,

More information

Training Restricted Boltzmann Machines on Word Observations

Training Restricted Boltzmann Machines on Word Observations In this work, we directly address the scalability issues associated with large softmax visible units in RBMs. We describe a learning rule with a computational complexity independent of the number of visible

More information

The connection of dropout and Bayesian statistics

The connection of dropout and Bayesian statistics The connection of dropout and Bayesian statistics Interpretation of dropout as approximate Bayesian modelling of NN http://mlg.eng.cam.ac.uk/yarin/thesis/thesis.pdf Dropout Geoffrey Hinton Google, University

More information

Learning Deep Architectures

Learning Deep Architectures Learning Deep Architectures Yoshua Bengio, U. Montreal Microsoft Cambridge, U.K. July 7th, 2009, Montreal Thanks to: Aaron Courville, Pascal Vincent, Dumitru Erhan, Olivier Delalleau, Olivier Breuleux,

More information

Knowledge Extraction from DBNs for Images

Knowledge Extraction from DBNs for Images Knowledge Extraction from DBNs for Images Son N. Tran and Artur d Avila Garcez Department of Computer Science City University London Contents 1 Introduction 2 Knowledge Extraction from DBNs 3 Experimental

More information

Learning and Evaluating Boltzmann Machines

Learning and Evaluating Boltzmann Machines Department of Computer Science 6 King s College Rd, Toronto University of Toronto M5S 3G4, Canada http://learning.cs.toronto.edu fax: +1 416 978 1455 Copyright c Ruslan Salakhutdinov 2008. June 26, 2008

More information

Representational Power of Restricted Boltzmann Machines and Deep Belief Networks. Nicolas Le Roux and Yoshua Bengio Presented by Colin Graber

Representational Power of Restricted Boltzmann Machines and Deep Belief Networks. Nicolas Le Roux and Yoshua Bengio Presented by Colin Graber Representational Power of Restricted Boltzmann Machines and Deep Belief Networks Nicolas Le Roux and Yoshua Bengio Presented by Colin Graber Introduction Representational abilities of functions with some

More information

Deep Learning. What Is Deep Learning? The Rise of Deep Learning. Long History (in Hind Sight)

Deep Learning. What Is Deep Learning? The Rise of Deep Learning. Long History (in Hind Sight) CSCE 636 Neural Networks Instructor: Yoonsuck Choe Deep Learning What Is Deep Learning? Learning higher level abstractions/representations from data. Motivation: how the brain represents sensory information

More information

Training an RBM: Contrastive Divergence. Sargur N. Srihari

Training an RBM: Contrastive Divergence. Sargur N. Srihari Training an RBM: Contrastive Divergence Sargur N. srihari@cedar.buffalo.edu Topics in Partition Function Definition of Partition Function 1. The log-likelihood gradient 2. Stochastic axiu likelihood and

More information

Robust Classification using Boltzmann machines by Vasileios Vasilakakis

Robust Classification using Boltzmann machines by Vasileios Vasilakakis Robust Classification using Boltzmann machines by Vasileios Vasilakakis The scope of this report is to propose an architecture of Boltzmann machines that could be used in the context of classification,

More information

Deep Belief Networks are Compact Universal Approximators

Deep Belief Networks are Compact Universal Approximators Deep Belief Networks are Compact Universal Approximators Franck Olivier Ndjakou Njeunje Applied Mathematics and Scientific Computation May 16, 2016 1 / 29 Outline 1 Introduction 2 Preliminaries Universal

More information

arxiv: v2 [cs.lg] 8 Sep 2015

arxiv: v2 [cs.lg] 8 Sep 2015 Sampled Weighted Min-Hashing for Large-Scale Topic Mining Gibran Fuentes-Pineda and Ivan Vladimir Meza-Ruiz Instituto de Investigaciones en Matemáticas y en Sistemas Universidad Nacional Autónoma de México

More information

Deep Belief Networks are compact universal approximators

Deep Belief Networks are compact universal approximators 1 Deep Belief Networks are compact universal approximators Nicolas Le Roux 1, Yoshua Bengio 2 1 Microsoft Research Cambridge 2 University of Montreal Keywords: Deep Belief Networks, Universal Approximation

More information

Empirical Analysis of the Divergence of Gibbs Sampling Based Learning Algorithms for Restricted Boltzmann Machines

Empirical Analysis of the Divergence of Gibbs Sampling Based Learning Algorithms for Restricted Boltzmann Machines Empirical Analysis of the Divergence of Gibbs Sampling Based Learning Algorithms for Restricted Boltzmann Machines Asja Fischer and Christian Igel Institut für Neuroinformatik Ruhr-Universität Bochum,

More information

Deep Learning & Neural Networks Lecture 2

Deep Learning & Neural Networks Lecture 2 Deep Learning & Neural Networks Lecture 2 Kevin Duh Graduate School of Information Science Nara Institute of Science and Technology Jan 16, 2014 2/45 Today s Topics 1 General Ideas in Deep Learning Motivation

More information

Classification of Higgs Boson Tau-Tau decays using GPU accelerated Neural Networks

Classification of Higgs Boson Tau-Tau decays using GPU accelerated Neural Networks Classification of Higgs Boson Tau-Tau decays using GPU accelerated Neural Networks Mohit Shridhar Stanford University mohits@stanford.edu, mohit@u.nus.edu Abstract In particle physics, Higgs Boson to tau-tau

More information

Contrastive Divergence

Contrastive Divergence Contrastive Divergence Training Products of Experts by Minimizing CD Hinton, 2002 Helmut Puhr Institute for Theoretical Computer Science TU Graz June 9, 2010 Contents 1 Theory 2 Argument 3 Contrastive

More information

Joint Training of Partially-Directed Deep Boltzmann Machines

Joint Training of Partially-Directed Deep Boltzmann Machines Joint Training of Partially-Directed Deep Boltzmann Machines Ian J. Goodfellow goodfeli@iro.umontreal.ca Aaron Courville aaron.courville@umontreal.ca Yoshua Bengio Département d Informatique et de Recherche

More information

Deep Neural Networks

Deep Neural Networks Deep Neural Networks DT2118 Speech and Speaker Recognition Giampiero Salvi KTH/CSC/TMH giampi@kth.se VT 2015 1 / 45 Outline State-to-Output Probability Model Artificial Neural Networks Perceptron Multi

More information

Distilling Model Knowledge

Distilling Model Knowledge University of Edinburgh School of Informatics MSc by Research in Data Science arxiv:1510.02437v1 [stat.ml] 8 Oct 2015 Distilling Model Knowledge by George Papamakarios August 2015 Abstract Top-performing

More information

J. Sadeghi E. Patelli M. de Angelis

J. Sadeghi E. Patelli M. de Angelis J. Sadeghi E. Patelli Institute for Risk and, Department of Engineering, University of Liverpool, United Kingdom 8th International Workshop on Reliable Computing, Computing with Confidence University of

More information

Connections between score matching, contrastive divergence, and pseudolikelihood for continuous-valued variables. Revised submission to IEEE TNN

Connections between score matching, contrastive divergence, and pseudolikelihood for continuous-valued variables. Revised submission to IEEE TNN Connections between score matching, contrastive divergence, and pseudolikelihood for continuous-valued variables Revised submission to IEEE TNN Aapo Hyvärinen Dept of Computer Science and HIIT University

More information

arxiv: v2 [cs.lg] 17 Nov 2016

arxiv: v2 [cs.lg] 17 Nov 2016 Approximating Wisdom of Crowds using K-RBMs Abhay Gupta Microsoft India R&D Pvt. Ltd. abhgup@microsoft.com arxiv:1611.05340v2 [cs.lg] 17 Nov 2016 Abstract An important way to make large training sets is

More information

How to do backpropagation in a brain. Geoffrey Hinton Canadian Institute for Advanced Research & University of Toronto

How to do backpropagation in a brain. Geoffrey Hinton Canadian Institute for Advanced Research & University of Toronto 1 How to do backpropagation in a brain Geoffrey Hinton Canadian Institute for Advanced Research & University of Toronto What is wrong with back-propagation? It requires labeled training data. (fixed) Almost

More information

Evaluation Methods for Topic Models

Evaluation Methods for Topic Models University of Massachusetts Amherst wallach@cs.umass.edu April 13, 2009 Joint work with Iain Murray, Ruslan Salakhutdinov and David Mimno Statistical Topic Models Useful for analyzing large, unstructured

More information

TUTORIAL PART 1 Unsupervised Learning

TUTORIAL PART 1 Unsupervised Learning TUTORIAL PART 1 Unsupervised Learning Marc'Aurelio Ranzato Department of Computer Science Univ. of Toronto ranzato@cs.toronto.edu Co-organizers: Honglak Lee, Yoshua Bengio, Geoff Hinton, Yann LeCun, Andrew

More information

Enhanced Gradient and Adaptive Learning Rate for Training Restricted Boltzmann Machines

Enhanced Gradient and Adaptive Learning Rate for Training Restricted Boltzmann Machines Enhanced Gradient and Adaptive Learning Rate for Training Restricted Boltzmann Machines KyungHyun Cho KYUNGHYUN.CHO@AALTO.FI Tapani Raiko TAPANI.RAIKO@AALTO.FI Alexander Ilin ALEXANDER.ILIN@AALTO.FI Department

More information

A Practical Guide to Training Restricted Boltzmann Machines

A Practical Guide to Training Restricted Boltzmann Machines Department of Computer Science 6 King s College Rd, Toronto University of Toronto M5S 3G4, Canada http://learning.cs.toronto.edu fax: +1 416 978 1455 Copyright c Geoffrey Hinton 2010. August 2, 2010 UTML

More information

Usually the estimation of the partition function is intractable and it becomes exponentially hard when the complexity of the model increases. However,

Usually the estimation of the partition function is intractable and it becomes exponentially hard when the complexity of the model increases. However, Odyssey 2012 The Speaker and Language Recognition Workshop 25-28 June 2012, Singapore First attempt of Boltzmann Machines for Speaker Verification Mohammed Senoussaoui 1,2, Najim Dehak 3, Patrick Kenny

More information

Deep Learning. What Is Deep Learning? The Rise of Deep Learning. Long History (in Hind Sight)

Deep Learning. What Is Deep Learning? The Rise of Deep Learning. Long History (in Hind Sight) CSCE 636 Neural Networks Instructor: Yoonsuck Choe Deep Learning What Is Deep Learning? Learning higher level abstractions/representations from data. Motivation: how the brain represents sensory information

More information

Implementation of a Restricted Boltzmann Machine in a Spiking Neural Network

Implementation of a Restricted Boltzmann Machine in a Spiking Neural Network Implementation of a Restricted Boltzmann Machine in a Spiking Neural Network Srinjoy Das Department of Electrical and Computer Engineering University of California, San Diego srinjoyd@gmail.com Bruno Umbria

More information

Deep Learning for Natural Language Processing

Deep Learning for Natural Language Processing Deep Learning for Natural Language Processing Dylan Drover, Borui Ye, Jie Peng University of Waterloo djdrover@uwaterloo.ca borui.ye@uwaterloo.ca July 8, 2015 Dylan Drover, Borui Ye, Jie Peng (University

More information

arxiv: v3 [cs.lg] 18 Mar 2013

arxiv: v3 [cs.lg] 18 Mar 2013 Hierarchical Data Representation Model - Multi-layer NMF arxiv:1301.6316v3 [cs.lg] 18 Mar 2013 Hyun Ah Song Department of Electrical Engineering KAIST Daejeon, 305-701 hyunahsong@kaist.ac.kr Abstract Soo-Young

More information

Inductive Principles for Restricted Boltzmann Machine Learning

Inductive Principles for Restricted Boltzmann Machine Learning Inductive Principles for Restricted Boltzmann Machine Learning Benjamin Marlin Department of Computer Science University of British Columbia Joint work with Kevin Swersky, Bo Chen and Nando de Freitas

More information

Deep Learning the Ising Model Near Criticality

Deep Learning the Ising Model Near Criticality Deep Learning the Ising Model ear Criticality It is empirically well supported that neural networks with deep architectures perform better than shallow networks for certain machine learning tasks involving

More information

Bayesian Learning in Undirected Graphical Models

Bayesian Learning in Undirected Graphical Models Bayesian Learning in Undirected Graphical Models Zoubin Ghahramani Gatsby Computational Neuroscience Unit University College London, UK http://www.gatsby.ucl.ac.uk/ Work with: Iain Murray and Hyun-Chul

More information

Efficient Learning of Sparse, Distributed, Convolutional Feature Representations for Object Recognition

Efficient Learning of Sparse, Distributed, Convolutional Feature Representations for Object Recognition Efficient Learning of Sparse, Distributed, Convolutional Feature Representations for Object Recognition Kihyuk Sohn Dae Yon Jung Honglak Lee Alfred O. Hero III Dept. of Electrical Engineering and Computer

More information

Spectral Hashing: Learning to Leverage 80 Million Images

Spectral Hashing: Learning to Leverage 80 Million Images Spectral Hashing: Learning to Leverage 80 Million Images Yair Weiss, Antonio Torralba, Rob Fergus Hebrew University, MIT, NYU Outline Motivation: Brute Force Computer Vision. Semantic Hashing. Spectral

More information

How to do backpropagation in a brain

How to do backpropagation in a brain How to do backpropagation in a brain Geoffrey Hinton Canadian Institute for Advanced Research & University of Toronto & Google Inc. Prelude I will start with three slides explaining a popular type of deep

More information

arxiv: v2 [cs.ne] 22 Feb 2013

arxiv: v2 [cs.ne] 22 Feb 2013 Sparse Penalty in Deep Belief Networks: Using the Mixed Norm Constraint arxiv:1301.3533v2 [cs.ne] 22 Feb 2013 Xanadu C. Halkias DYNI, LSIS, Universitè du Sud, Avenue de l Université - BP20132, 83957 LA

More information

CSC321 Lecture 20: Autoencoders

CSC321 Lecture 20: Autoencoders CSC321 Lecture 20: Autoencoders Roger Grosse Roger Grosse CSC321 Lecture 20: Autoencoders 1 / 16 Overview Latent variable models so far: mixture models Boltzmann machines Both of these involve discrete

More information

Scaling Neighbourhood Methods

Scaling Neighbourhood Methods Quick Recap Scaling Neighbourhood Methods Collaborative Filtering m = #items n = #users Complexity : m * m * n Comparative Scale of Signals ~50 M users ~25 M items Explicit Ratings ~ O(1M) (1 per billion)

More information

Gaussian-Bernoulli Deep Boltzmann Machine

Gaussian-Bernoulli Deep Boltzmann Machine Gaussian-Bernoulli Deep Boltzmann Machine KyungHyun Cho, Tapani Raiko and Alexander Ilin Aalto University School of Science Department of Information and Computer Science Espoo, Finland {kyunghyun.cho,tapani.raiko,alexander.ilin}@aalto.fi

More information

Basic Principles of Unsupervised and Unsupervised

Basic Principles of Unsupervised and Unsupervised Basic Principles of Unsupervised and Unsupervised Learning Toward Deep Learning Shun ichi Amari (RIKEN Brain Science Institute) collaborators: R. Karakida, M. Okada (U. Tokyo) Deep Learning Self Organization

More information

LEARNING SPARSE STRUCTURED ENSEMBLES WITH STOCASTIC GTADIENT MCMC SAMPLING AND NETWORK PRUNING

LEARNING SPARSE STRUCTURED ENSEMBLES WITH STOCASTIC GTADIENT MCMC SAMPLING AND NETWORK PRUNING LEARNING SPARSE STRUCTURED ENSEMBLES WITH STOCASTIC GTADIENT MCMC SAMPLING AND NETWORK PRUNING Yichi Zhang Zhijian Ou Speech Processing and Machine Intelligence (SPMI) Lab Department of Electronic Engineering

More information

An overview of word2vec

An overview of word2vec An overview of word2vec Benjamin Wilson Berlin ML Meetup, July 8 2014 Benjamin Wilson word2vec Berlin ML Meetup 1 / 25 Outline 1 Introduction 2 Background & Significance 3 Architecture 4 CBOW word representations

More information

Deep Learning Autoencoder Models

Deep Learning Autoencoder Models Deep Learning Autoencoder Models Davide Bacciu Dipartimento di Informatica Università di Pisa Intelligent Systems for Pattern Recognition (ISPR) Generative Models Wrap-up Deep Learning Module Lecture Generative

More information

A Unified View of Deep Generative Models

A Unified View of Deep Generative Models SAILING LAB Laboratory for Statistical Artificial InteLigence & INtegreative Genomics A Unified View of Deep Generative Models Zhiting Hu and Eric Xing Petuum Inc. Carnegie Mellon University 1 Deep generative

More information

Measuring the Usefulness of Hidden Units in Boltzmann Machines with Mutual Information

Measuring the Usefulness of Hidden Units in Boltzmann Machines with Mutual Information Measuring the Usefulness of Hidden Units in Boltzmann Machines with Mutual Information Mathias Berglund, Tapani Raiko, and KyungHyun Cho Department of Information and Computer Science Aalto University

More information

Notes on Boltzmann Machines

Notes on Boltzmann Machines 1 Notes on Boltzmann Machines Patrick Kenny Centre de recherche informatique de Montréal Patrick.Kenny@crim.ca I. INTRODUCTION Boltzmann machines are probability distributions on high dimensional binary

More information

Cardinality Restricted Boltzmann Machines

Cardinality Restricted Boltzmann Machines Cardinality Restricted Boltzmann Machines Kevin Swersky Daniel Tarlow Ilya Sutskever Dept. of Computer Science University of Toronto [kswersky,dtarlow,ilya]@cs.toronto.edu Ruslan Salakhutdinov, Richard

More information

Neural Networks: Backpropagation

Neural Networks: Backpropagation Neural Networks: Backpropagation Seung-Hoon Na 1 1 Department of Computer Science Chonbuk National University 2018.10.25 eung-hoon Na (Chonbuk National University) Neural Networks: Backpropagation 2018.10.25

More information

Latent variable models for discrete data

Latent variable models for discrete data Latent variable models for discrete data Jianfei Chen Department of Computer Science and Technology Tsinghua University, Beijing 100084 chris.jianfei.chen@gmail.com Janurary 13, 2014 Murphy, Kevin P. Machine

More information

Chapter 11. Stochastic Methods Rooted in Statistical Mechanics

Chapter 11. Stochastic Methods Rooted in Statistical Mechanics Chapter 11. Stochastic Methods Rooted in Statistical Mechanics Neural Networks and Learning Machines (Haykin) Lecture Notes on Self-learning Neural Algorithms Byoung-Tak Zhang School of Computer Science

More information

Building a Multi-FPGA Virtualized Restricted Boltzmann Machine Architecture Using Embedded MPI

Building a Multi-FPGA Virtualized Restricted Boltzmann Machine Architecture Using Embedded MPI Building a Multi-FPGA Virtualized Restricted Boltzmann Machine Architecture Using Embedded MPI Charles Lo and Paul Chow {locharl1, pc}@eecg.toronto.edu Department of Electrical and Computer Engineering

More information

Data Mining & Machine Learning

Data Mining & Machine Learning Data Mining & Machine Learning CS57300 Purdue University March 1, 2018 1 Recap of Last Class (Model Search) Forward and Backward passes 2 Feedforward Neural Networks Neural Networks: Architectures 2-layer

More information

Index. Santanu Pattanayak 2017 S. Pattanayak, Pro Deep Learning with TensorFlow,

Index. Santanu Pattanayak 2017 S. Pattanayak, Pro Deep Learning with TensorFlow, Index A Activation functions, neuron/perceptron binary threshold activation function, 102 103 linear activation function, 102 rectified linear unit, 106 sigmoid activation function, 103 104 SoftMax activation

More information