Adaptive Dropout Training for SVMs

Size: px
Start display at page:

Download "Adaptive Dropout Training for SVMs"

Transcription

1 Department of Computer Science and Technology Adaptive Dropout Training for SVMs Jun Zhu Joint with Ning Chen, Jingwei Zhuo, Jianfei Chen, Bo Zhang Tsinghua University ShanghaiTech Symposium on Data Science, June 23-26, 2015

2 Outline Overfitting in Big Data Dropout training for SVMs Adaptive dropout rates Big learning with Bayesian methods Conclusions 1

3 Overfitting Bias-variance tradeoff risk variance / estimation error bias / approximation error complexity of function class 2

4 Overfitting in Big Data Big Model + Big Data + Big/Super Cluster Big ML 9 layers sparse autoencoder with: -local receptive fields to scale up; - local L2 pooling and local contrast normalization for invariant features - 1B parameters (connections) - 10M 200x200 images - train with 1K machines (16K cores) for 3 days [Le et al., 2012] -able to build high-level concepts, e.g., cat faces and human bodies -15.8% accuracy in recognizing 22K objects (70% relative improvements) 3

5 Overfitting in Big Data Relevant information grows slower than linear Model capacity may grow faster than the amount of relevant information!

6 Overfitting in Big Data Relevant information grows slower than linear (Bialek et al., 2001)

7 Overfitting in Big Data Regularization to prevent overfitting is increasingly important, rather than increasingly irrelevant! Increasing attention, e.g., dropout training (Hinton, 2012) More theoretical understanding and extensions MCF (van der Maaten et al., 2013); Logistic-loss (Wager et al., 2013); Generalization error (Wager et al., 2014) Dropout SVM (Chen et al., 2014); Adaptive Dropout (Zhuo et al., 2015) 6

8 Amazon Reviews Classification Positive or Negative? I love the deeper meaning behind this movie. I had watched it years ago when it first came out but never understood it til now. Great spillberg film One of the best sci-fi /adventure movies I have ever seen. Great movie about robots and ones yearning to know ones creator. The ending will stick in your mind forever in a good way. a massive, manipulative tear jerker which did nothing to illuminate me on the subjects of love or parenting or relationship or science or sibling rivalry. stock characters Very long. Boring stretches made this movie hard to finish 7

9 Amazon Reviews Classification Positive or Negative? Regularized Empirical Risk Minimization: Instead of regularizing parameters, we incorporate knowledge directly from data to do regularization? Regularization by corrupting data 8

10 Regularization by Corruptions Original Features Corrupted Features Corrupted Features I love the deeper meaning behind this movie. I had watched it years ago when it first came out but never understood it til now. Great spillberg film One of the best sci-fi /adventure movies I have ever seen. Great movie about robots and ones yearning to know ones creator. The ending will stick in your mind forever in a good way. a massive, manipulative tear jerker which did nothing to illuminate me on the subjects of love or parenting or relationship or science or sibling rivalry. stock characters I love the deeper meaning behind this movie. I had watched it years ago when it first came out but never understood it til now. Great spillberg film One of the best sci-fi /adventure movies I have ever seen. Great movie about robots and ones yearning to know ones creator. The ending will stick in your mind forever in a good way. a massive, manipulative tear jerker which did nothing to illuminate me on the subjects of love or parenting or relationship or science or sibling rivalry. stock characters I love the deeper meaning behind this movie. I had watched it years ago when it first came out but never understood it til now. Great spillberg film One of the best sci-fi /adventure movies I have ever seen. Great movie about robots and ones yearning to know ones creator. The ending will stick in your mind forever in a good way. a massive, manipulative tear jerker which did nothing to illuminate me on the subjects of love or parenting or relationship or science or sibling rivalry. stock characters Very long. Boring stretches made this movie hard to finish Very long. Boring stretches made this movie hard to finish Very long. Boring stretches made this movie hard to finish 9

11 Feature Noising Models Define a label-invariant corrupting distribution, assume the corruption is independent across features Very long. Boring stretches made this movie hard to finish x x d Very long. Boring stretches made this movie hard to finish ~x d ~x d D Corrupting distributions: Dropout Gaussian Laplace Poisson 10

12 Feature Noising Models Feature noising models control over-fitting! (Hinton et.al, 2012) Augment the data size by corrupting the given training examples with a fixed noise distribution. Very long. Boring stretches made this movie hard to finish Very long. Boring stretches made this movie hard to finish Explicit Corruption Downside: gets computationally prohibitive, unless... 11

13 Feature Noising Models Feature noising models control over-fitting! (Hinton et.al, 2012) Augment the data size by corrupting the given training examples with a fixed noise distribution. Very long. Boring stretches made this movie hard to finish Very long. Boring stretches made this movie hard to finish Explicit Corruption Downside: gets computationally prohibitive, unless Implicit Corruption (MCF) 12

14 Regularization by corrupting data Feature noising as minimizing the expected loss functions Theoretical understanding L2-regularization for additive Gaussian noise (Bishop, 1995) Adaptive regularization for dropout LR (Wager et. al., 2013) Generalization bound (Wager et al., 2014) Empirical results in various applications Document classification (van der Maaten et. al., 2013); Entity recognition (Wang et.al., 2013); Image classification (Wang & Manning, 2013). Expected Losses Quadratic loss (Bishop, 1995) Exponential loss (van der Maaten et. al., 2013) Logistic loss (van der Maaten et. al., 2013; Wager et al., 2013) 13

15 Losses in Machine Learning 14

16 Dropout Training for Support Vector Machines Explicit corruption for SVM (Burges & Scholkopf, 1997) One technical challenge is the non-smoothness of the hinge loss makes it hard to compute Intractable! Our work: Develop an iteratively re-weighted least square (IRLS) algorithm to minimize a variational bound; Apply ideas to develop IRLS for dropout Logistic regression; Derive an adaptive learning rule to decide the noise level [Chen et al., AAAI 2014; Zhuo et al., IJCAI 2015] 15

17 16 Variational bound with Data Augmentation Theorem: Let, and be the pseudo-likelihood of response variables for sample n. The pseudo-likelihood can be expressed as a scale-location mixture of Gaussians where is a generalized inverse Gaussian variable, proof follows (Polson & Scott, 2011) with some careful treatments.

18 Variational bound with Data Augmentation A variational bound with data augmentation The expected hinge loss is: (c is the regularization parameter) Using the ideas of data augmentation Variational upper bound 17

19 Iteratively Re-weighted Least Square Algorithm Variational optimization problem Coordinate Descent (variational EM) For (i.e., E-step) 18

20 Iteratively Re-weighted Least Square Algorithm Coordinate Descent (variational EM) For (i.e., M-step) A re-weighted least square problem under feature noising adaptive weights h n := E q [ 1 n ]= Reduces to the square loss by setting 1 c p E p [³ 2 n] re-weighted label yn h = (`+ 1 )y n c n h n = 1 c ; and ` = 0 19

21 20 Variational bound for Expected Logistic Loss The expected logistic loss function Theorem: Let, and be the pseudo-likelihood of response variables where, and is a Polya-Gamma variable, proof follows (Polson & Scott, 2012) with some careful treatments.

22 Variational bound for Expected Logistic Loss Variational optimization problem Coordinate Descent (variational EM) For (i.e., E-step) a Polya-Gamma distribution (Polson et al., 2012) 21

23 Iteratively Re-weighted Least Square Algorithm Coordinate Descent (variational EM) For (i.e., M-step) A re-weighted least square problem under feature noising l n := 1 c E q[ n ]= adaptive weights Reduces to the square loss by setting 1 2 p E p [! n 2 ] 1 + e ep E p [! 2 n ] 1 p E p [! 2 n ] l n = c 2 re-weighted label y l n = c 2 n y n 22

24 23 Comparison of logistic / hinge loss under IRLS Comparison of hinge & logistic loss under the IRLS framework. Hinge Parameter Parameter c Update Update Logistic -- c ` ` n y n c n h yn h n l yn l Both losses iteratively minimize the expectation of a reweighted quadratic loss, but differ in the update rules of the weights and the labels at each iteration. Quadratic loss is a special case with a single iteration

25 Experiment Compare Dropout-SVM and Dropout-Logistic with state-ofart models: MCF-logistic, MCF-quadratic All our predictors use L2-regularization, with parameters set by cross-validation Single-parameter noising model: x d x d d ~x d D ~x d D 24

26 Better 25 Experiment 1: Review Classification (P / N) No Corruption No Corruption No Corruption No Corruption

27 Experiment 1: Review Classification (P / N) Comparing explicit and implicit dropout corruption (Amazon Books; hinge loss): 26

28 Experiment 2: Nightmare at test time In some settings, features may be randomly unobserved at test time We experiment with this nightmare at test time scenario on MNIST digits: Train regular dropout classifiers on the original training set Randomly delete features from the test images, and measure classification error 27

29 Better 28 Experiment 2: Nightmare at test time Classification error on test images with randomly deleted features: More test corruption

30 Adaptive Dropout Rates A Bayesian feature noising model p( i j ) = ~x i = x i ± i Y D d= 1 (1 d) id 1 id x id ~x id Allows various dimensions to have different dropout rates d Automatically infer the dropout rates (non-informative prior) d id D d N ^ d = P N i= 1 I(y iµ d x id < 0) P N i= 1 I(x id 6= 0) Group-wise structure among features is allowed [Zhuo, Zhu, & Zhang, 2015] 29

31 30 Adaptive Dropout Rates Some results on Amazon books and kitchen review data: Adaptive rates can improve the performance

32 Big Learning with Bayesian Methods Why Bayes? Robust to overfitting Flexible in modeling Avoid (heavy) parameter tuning Generic algorithms to do inference Why not Bayes? Computationally too slow Not scalable to big data Good news: Much recent progress on scalable Bayesian methods 31

33 Big Learning with Bayesian Methods Stochastic/Online Methods Variational, MCMC Distributed Methods Variational, MCMC Data-Parallel Graph-Parallel Model-Parallel master server map reduce slave client 32

34 Big Learning with Bayesian Methods Online/Stochastic Learning Online Bayesian PA (Shi & Zhu, ICML 2014) Stochastic subgradient MCMC (Hu et al., arxiv: , 2015) Deep generative models (Li et al., arxiv: , 2015; Du et al., arxiv: , 2015) 33

35 Big Learning with Bayesian Methods Online/Stochastic Learning Online Bayesian PA (Shi & Zhu, ICML 2014) Stochastic subgradient MCMC (Hu et al., arxiv: , 2015) Deep generative models (Li et al., arxiv: , 2015; Du et al., arxiv: , 2015) 34

36 Big Learning with Bayesian Methods Online/Stochastic Learning Online Bayesian PA (Shi & Zhu, ICML 2014) Stochastic subgradient MCMC (Hu et al., arxiv: , 2015) Deep generative models (Li et al., arxiv: , 2015; Du et al., arxiv: , 2015) 35

37 Big Learning with Bayesian Methods Online/Stochastic Learning Online Bayesian PA (Shi & Zhu, ICML 2014) Stochastic subgradient MCMC (Hu et al., arxiv: , 2015) Deep generative models (Li et al., arxiv: , 2015; Du et al., arxiv: , 2015) Distributed Learning Distributed Bayesian Inference (Xu et al., NIPS 2014) Scalable topic graph learning (Chen et al., NIPS 2013) Scalable dynamic LDA (Bhadury et al., 2015, preprint) A comprehensive survey Big Learning with Bayesian Methods, Zhu et al., arxiv:

38 Conclusions Feature noising controls over-fitting Dropout training for SVMs with an iteratively re-weighted least square (IRLS) algorithm Apply ideas to develop IRLS for dropout Logistic regression Adaptive update rule for dropout levels Future Work Kernel trick in dropout learning Dropout-SVM in deep architectures Big learning with Bayesian methods 37

39 Department of Computer Science and Technology Thank you!

Adaptive Dropout Rates for Learning with Corrupted Features

Adaptive Dropout Rates for Learning with Corrupted Features Adaptive Dropout Rates for Learning with Corrupted Features Jingwei Zhuo, Jun Zhu, Bo Zhang Dept. of Comp. Sci. & Tech., State Key Lab of Intell. Tech. & Sys., TNList Lab, Center for Bio-Inspired Computing

More information

Adaptive Dropout Rates for Learning with Corrupted Features

Adaptive Dropout Rates for Learning with Corrupted Features Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence (IJCAI 2015) Adaptive Dropout Rates for Learning with Corrupted Features Jingwei Zhuo, Jun Zhu, Bo Zhang Dept.

More information

Deep Learning With Noise

Deep Learning With Noise Deep Learning With Noise Yixin Luo Computer Science Department Carnegie Mellon University yixinluo@cs.cmu.edu Fan Yang Department of Mathematical Sciences Carnegie Mellon University fanyang1@andrew.cmu.edu

More information

Deep Generative Models and a Probabilistic Programming Library

Deep Generative Models and a Probabilistic Programming Library Deep Generative Models and a Probabilistic Programming Library Discriminative (Deep) Learning Learn a (differentiable) function mapping from input to output x f(x; θ) y Gradient back-propagation Generative

More information

Classification: Linear Discriminant Functions

Classification: Linear Discriminant Functions Classification: Linear Discriminant Functions CE-725: Statistical Pattern Recognition Sharif University of Technology Spring 2013 Soleymani Outline Discriminant functions Linear Discriminant functions

More information

ECS289: Scalable Machine Learning

ECS289: Scalable Machine Learning ECS289: Scalable Machine Learning Cho-Jui Hsieh UC Davis Sept 22, 2016 Course Information Website: http://www.stat.ucdavis.edu/~chohsieh/teaching/ ECS289G_Fall2016/main.html My office: Mathematical Sciences

More information

Preface to the Second Edition. Preface to the First Edition. 1 Introduction 1

Preface to the Second Edition. Preface to the First Edition. 1 Introduction 1 Preface to the Second Edition Preface to the First Edition vii xi 1 Introduction 1 2 Overview of Supervised Learning 9 2.1 Introduction... 9 2.2 Variable Types and Terminology... 9 2.3 Two Simple Approaches

More information

Learning with Marginalized Corrupted Features and Labels Together

Learning with Marginalized Corrupted Features and Labels Together Learning with Marginalized Corrupted Features and Labels Together Yingming Li, Ming Yang, Zenglin Xu, and Zhongfei (Mark) Zhang School of Computer Science and Engineering, Big Data Research Center University

More information

Structured Learning. Jun Zhu

Structured Learning. Jun Zhu Structured Learning Jun Zhu Supervised learning Given a set of I.I.D. training samples Learn a prediction function b r a c e Supervised learning (cont d) Many different choices Logistic Regression Maximum

More information

Machine Learning Basics: Stochastic Gradient Descent. Sargur N. Srihari

Machine Learning Basics: Stochastic Gradient Descent. Sargur N. Srihari Machine Learning Basics: Stochastic Gradient Descent Sargur N. srihari@cedar.buffalo.edu 1 Topics 1. Learning Algorithms 2. Capacity, Overfitting and Underfitting 3. Hyperparameters and Validation Sets

More information

COMP 551 Applied Machine Learning Lecture 16: Deep Learning

COMP 551 Applied Machine Learning Lecture 16: Deep Learning COMP 551 Applied Machine Learning Lecture 16: Deep Learning Instructor: Ryan Lowe (ryan.lowe@cs.mcgill.ca) Slides mostly by: Class web page: www.cs.mcgill.ca/~hvanho2/comp551 Unless otherwise noted, all

More information

Machine Learning Lecture 3

Machine Learning Lecture 3 Machine Learning Lecture 3 Probability Density Estimation II 19.10.2017 Bastian Leibe RWTH Aachen http://www.vision.rwth-aachen.de leibe@vision.rwth-aachen.de Announcements Exam dates We re in the process

More information

FMA901F: Machine Learning Lecture 3: Linear Models for Regression. Cristian Sminchisescu

FMA901F: Machine Learning Lecture 3: Linear Models for Regression. Cristian Sminchisescu FMA901F: Machine Learning Lecture 3: Linear Models for Regression Cristian Sminchisescu Machine Learning: Frequentist vs. Bayesian In the frequentist setting, we seek a fixed parameter (vector), with value(s)

More information

Analysis: TextonBoost and Semantic Texton Forests. Daniel Munoz Februrary 9, 2009

Analysis: TextonBoost and Semantic Texton Forests. Daniel Munoz Februrary 9, 2009 Analysis: TextonBoost and Semantic Texton Forests Daniel Munoz 16-721 Februrary 9, 2009 Papers [shotton-eccv-06] J. Shotton, J. Winn, C. Rother, A. Criminisi, TextonBoost: Joint Appearance, Shape and Context

More information

Case Study 1: Estimating Click Probabilities

Case Study 1: Estimating Click Probabilities Case Study 1: Estimating Click Probabilities SGD cont d AdaGrad Machine Learning for Big Data CSE547/STAT548, University of Washington Sham Kakade March 31, 2015 1 Support/Resources Office Hours Yao Lu:

More information

ImageNet Classification with Deep Convolutional Neural Networks

ImageNet Classification with Deep Convolutional Neural Networks ImageNet Classification with Deep Convolutional Neural Networks Alex Krizhevsky Ilya Sutskever Geoffrey Hinton University of Toronto Canada Paper with same name to appear in NIPS 2012 Main idea Architecture

More information

Variational Methods for Discrete-Data Latent Gaussian Models

Variational Methods for Discrete-Data Latent Gaussian Models Variational Methods for Discrete-Data Latent Gaussian Models University of British Columbia Vancouver, Canada March 6, 2012 The Big Picture Joint density models for data with mixed data types Bayesian

More information

Homework. Gaussian, Bishop 2.3 Non-parametric, Bishop 2.5 Linear regression Pod-cast lecture on-line. Next lectures:

Homework. Gaussian, Bishop 2.3 Non-parametric, Bishop 2.5 Linear regression Pod-cast lecture on-line. Next lectures: Homework Gaussian, Bishop 2.3 Non-parametric, Bishop 2.5 Linear regression 3.0-3.2 Pod-cast lecture on-line Next lectures: I posted a rough plan. It is flexible though so please come with suggestions Bayes

More information

Boosting Simple Model Selection Cross Validation Regularization

Boosting Simple Model Selection Cross Validation Regularization Boosting: (Linked from class website) Schapire 01 Boosting Simple Model Selection Cross Validation Regularization Machine Learning 10701/15781 Carlos Guestrin Carnegie Mellon University February 8 th,

More information

Neural Networks. Single-layer neural network. CSE 446: Machine Learning Emily Fox University of Washington March 10, /10/2017

Neural Networks. Single-layer neural network. CSE 446: Machine Learning Emily Fox University of Washington March 10, /10/2017 3/0/207 Neural Networks Emily Fox University of Washington March 0, 207 Slides adapted from Ali Farhadi (via Carlos Guestrin and Luke Zettlemoyer) Single-layer neural network 3/0/207 Perceptron as a neural

More information

CS 229 Midterm Review

CS 229 Midterm Review CS 229 Midterm Review Course Staff Fall 2018 11/2/2018 Outline Today: SVMs Kernels Tree Ensembles EM Algorithm / Mixture Models [ Focus on building intuition, less so on solving specific problems. Ask

More information

Class 6 Large-Scale Image Classification

Class 6 Large-Scale Image Classification Class 6 Large-Scale Image Classification Liangliang Cao, March 7, 2013 EECS 6890 Topics in Information Processing Spring 2013, Columbia University http://rogerioferis.com/visualrecognitionandsearch Visual

More information

Building Classifiers using Bayesian Networks

Building Classifiers using Bayesian Networks Building Classifiers using Bayesian Networks Nir Friedman and Moises Goldszmidt 1997 Presented by Brian Collins and Lukas Seitlinger Paper Summary The Naive Bayes classifier has reasonable performance

More information

Deep Learning for Computer Vision II

Deep Learning for Computer Vision II IIIT Hyderabad Deep Learning for Computer Vision II C. V. Jawahar Paradigm Shift Feature Extraction (SIFT, HoG, ) Part Models / Encoding Classifier Sparrow Feature Learning Classifier Sparrow L 1 L 2 L

More information

Supplementary material for: BO-HB: Robust and Efficient Hyperparameter Optimization at Scale

Supplementary material for: BO-HB: Robust and Efficient Hyperparameter Optimization at Scale Supplementary material for: BO-: Robust and Efficient Hyperparameter Optimization at Scale Stefan Falkner 1 Aaron Klein 1 Frank Hutter 1 A. Available Software To promote reproducible science and enable

More information

Unsupervised Learning

Unsupervised Learning Deep Learning for Graphics Unsupervised Learning Niloy Mitra Iasonas Kokkinos Paul Guerrero Vladimir Kim Kostas Rematas Tobias Ritschel UCL UCL/Facebook UCL Adobe Research U Washington UCL Timetable Niloy

More information

What is machine learning?

What is machine learning? Machine learning, pattern recognition and statistical data modelling Lecture 12. The last lecture Coryn Bailer-Jones 1 What is machine learning? Data description and interpretation finding simpler relationship

More information

Deep Learning for Computer Vision

Deep Learning for Computer Vision Deep Learning for Computer Vision Lecture 7: Universal Approximation Theorem, More Hidden Units, Multi-Class Classifiers, Softmax, and Regularization Peter Belhumeur Computer Science Columbia University

More information

Probabilistic Graphical Models

Probabilistic Graphical Models Overview of Part Two Probabilistic Graphical Models Part Two: Inference and Learning Christopher M. Bishop Exact inference and the junction tree MCMC Variational methods and EM Example General variational

More information

Bayesian model ensembling using meta-trained recurrent neural networks

Bayesian model ensembling using meta-trained recurrent neural networks Bayesian model ensembling using meta-trained recurrent neural networks Luca Ambrogioni l.ambrogioni@donders.ru.nl Umut Güçlü u.guclu@donders.ru.nl Yağmur Güçlütürk y.gucluturk@donders.ru.nl Julia Berezutskaya

More information

Neural Networks. CE-725: Statistical Pattern Recognition Sharif University of Technology Spring Soleymani

Neural Networks. CE-725: Statistical Pattern Recognition Sharif University of Technology Spring Soleymani Neural Networks CE-725: Statistical Pattern Recognition Sharif University of Technology Spring 2013 Soleymani Outline Biological and artificial neural networks Feed-forward neural networks Single layer

More information

More Data, Less Work: Runtime as a decreasing function of data set size. Nati Srebro. Toyota Technological Institute Chicago

More Data, Less Work: Runtime as a decreasing function of data set size. Nati Srebro. Toyota Technological Institute Chicago More Data, Less Work: Runtime as a decreasing function of data set size Nati Srebro Toyota Technological Institute Chicago Outline we are here SVM speculations, other problems Clustering wild speculations,

More information

Boosting Simple Model Selection Cross Validation Regularization. October 3 rd, 2007 Carlos Guestrin [Schapire, 1989]

Boosting Simple Model Selection Cross Validation Regularization. October 3 rd, 2007 Carlos Guestrin [Schapire, 1989] Boosting Simple Model Selection Cross Validation Regularization Machine Learning 10701/15781 Carlos Guestrin Carnegie Mellon University October 3 rd, 2007 1 Boosting [Schapire, 1989] Idea: given a weak

More information

Efficient Deep Learning Optimization Methods

Efficient Deep Learning Optimization Methods 11-785/ Spring 2019/ Recitation 3 Efficient Deep Learning Optimization Methods Josh Moavenzadeh, Kai Hu, and Cody Smith Outline 1 Review of optimization 2 Optimization practice 3 Training tips in PyTorch

More information

Challenges motivating deep learning. Sargur N. Srihari

Challenges motivating deep learning. Sargur N. Srihari Challenges motivating deep learning Sargur N. srihari@cedar.buffalo.edu 1 Topics In Machine Learning Basics 1. Learning Algorithms 2. Capacity, Overfitting and Underfitting 3. Hyperparameters and Validation

More information

Dynamic Bayesian network (DBN)

Dynamic Bayesian network (DBN) Readings: K&F: 18.1, 18.2, 18.3, 18.4 ynamic Bayesian Networks Beyond 10708 Graphical Models 10708 Carlos Guestrin Carnegie Mellon University ecember 1 st, 2006 1 ynamic Bayesian network (BN) HMM defined

More information

10601 Machine Learning. Model and feature selection

10601 Machine Learning. Model and feature selection 10601 Machine Learning Model and feature selection Model selection issues We have seen some of this before Selecting features (or basis functions) Logistic regression SVMs Selecting parameter value Prior

More information

Using Machine Learning to Optimize Storage Systems

Using Machine Learning to Optimize Storage Systems Using Machine Learning to Optimize Storage Systems Dr. Kiran Gunnam 1 Outline 1. Overview 2. Building Flash Models using Logistic Regression. 3. Storage Object classification 4. Storage Allocation recommendation

More information

Data Mining Practical Machine Learning Tools and Techniques. Slides for Chapter 6 of Data Mining by I. H. Witten and E. Frank

Data Mining Practical Machine Learning Tools and Techniques. Slides for Chapter 6 of Data Mining by I. H. Witten and E. Frank Data Mining Practical Machine Learning Tools and Techniques Slides for Chapter 6 of Data Mining by I. H. Witten and E. Frank Implementation: Real machine learning schemes Decision trees Classification

More information

Sum-Product Networks. STAT946 Deep Learning Guest Lecture by Pascal Poupart University of Waterloo October 15, 2015

Sum-Product Networks. STAT946 Deep Learning Guest Lecture by Pascal Poupart University of Waterloo October 15, 2015 Sum-Product Networks STAT946 Deep Learning Guest Lecture by Pascal Poupart University of Waterloo October 15, 2015 Introduction Outline What is a Sum-Product Network? Inference Applications In more depth

More information

A Fast Learning Algorithm for Deep Belief Nets

A Fast Learning Algorithm for Deep Belief Nets A Fast Learning Algorithm for Deep Belief Nets Geoffrey E. Hinton, Simon Osindero Department of Computer Science University of Toronto, Toronto, Canada Yee-Whye Teh Department of Computer Science National

More information

DS Machine Learning and Data Mining I. Alina Oprea Associate Professor, CCIS Northeastern University

DS Machine Learning and Data Mining I. Alina Oprea Associate Professor, CCIS Northeastern University DS 4400 Machine Learning and Data Mining I Alina Oprea Associate Professor, CCIS Northeastern University September 20 2018 Review Solution for multiple linear regression can be computed in closed form

More information

Recognition Tools: Support Vector Machines

Recognition Tools: Support Vector Machines CS 2770: Computer Vision Recognition Tools: Support Vector Machines Prof. Adriana Kovashka University of Pittsburgh January 12, 2017 Announcement TA office hours: Tuesday 4pm-6pm Wednesday 10am-12pm Matlab

More information

Applied Bayesian Nonparametrics 5. Spatial Models via Gaussian Processes, not MRFs Tutorial at CVPR 2012 Erik Sudderth Brown University

Applied Bayesian Nonparametrics 5. Spatial Models via Gaussian Processes, not MRFs Tutorial at CVPR 2012 Erik Sudderth Brown University Applied Bayesian Nonparametrics 5. Spatial Models via Gaussian Processes, not MRFs Tutorial at CVPR 2012 Erik Sudderth Brown University NIPS 2008: E. Sudderth & M. Jordan, Shared Segmentation of Natural

More information

Recap: Gaussian (or Normal) Distribution. Recap: Minimizing the Expected Loss. Topics of This Lecture. Recap: Maximum Likelihood Approach

Recap: Gaussian (or Normal) Distribution. Recap: Minimizing the Expected Loss. Topics of This Lecture. Recap: Maximum Likelihood Approach Truth Course Outline Machine Learning Lecture 3 Fundamentals (2 weeks) Bayes Decision Theory Probability Density Estimation Probability Density Estimation II 2.04.205 Discriminative Approaches (5 weeks)

More information

10-701/15-781, Fall 2006, Final

10-701/15-781, Fall 2006, Final -7/-78, Fall 6, Final Dec, :pm-8:pm There are 9 questions in this exam ( pages including this cover sheet). If you need more room to work out your answer to a question, use the back of the page and clearly

More information

Generative and discriminative classification techniques

Generative and discriminative classification techniques Generative and discriminative classification techniques Machine Learning and Category Representation 013-014 Jakob Verbeek, December 13+0, 013 Course website: http://lear.inrialpes.fr/~verbeek/mlcr.13.14

More information

Deep Learning. Vladimir Golkov Technical University of Munich Computer Vision Group

Deep Learning. Vladimir Golkov Technical University of Munich Computer Vision Group Deep Learning Vladimir Golkov Technical University of Munich Computer Vision Group 1D Input, 1D Output target input 2 2D Input, 1D Output: Data Distribution Complexity Imagine many dimensions (data occupies

More information

Machine Learning / Jan 27, 2010

Machine Learning / Jan 27, 2010 Revisiting Logistic Regression & Naïve Bayes Aarti Singh Machine Learning 10-701/15-781 Jan 27, 2010 Generative and Discriminative Classifiers Training classifiers involves learning a mapping f: X -> Y,

More information

Conditional Random Fields and beyond D A N I E L K H A S H A B I C S U I U C,

Conditional Random Fields and beyond D A N I E L K H A S H A B I C S U I U C, Conditional Random Fields and beyond D A N I E L K H A S H A B I C S 5 4 6 U I U C, 2 0 1 3 Outline Modeling Inference Training Applications Outline Modeling Problem definition Discriminative vs. Generative

More information

CS6220: DATA MINING TECHNIQUES

CS6220: DATA MINING TECHNIQUES CS6220: DATA MINING TECHNIQUES Image Data: Classification via Neural Networks Instructor: Yizhou Sun yzsun@ccs.neu.edu November 19, 2015 Methods to Learn Classification Clustering Frequent Pattern Mining

More information

Matrix Computations and " Neural Networks in Spark

Matrix Computations and  Neural Networks in Spark Matrix Computations and " Neural Networks in Spark Reza Zadeh Paper: http://arxiv.org/abs/1509.02256 Joint work with many folks on paper. @Reza_Zadeh http://reza-zadeh.com Training Neural Networks Datasets

More information

Simple Model Selection Cross Validation Regularization Neural Networks

Simple Model Selection Cross Validation Regularization Neural Networks Neural Nets: Many possible refs e.g., Mitchell Chapter 4 Simple Model Selection Cross Validation Regularization Neural Networks Machine Learning 10701/15781 Carlos Guestrin Carnegie Mellon University February

More information

JOINT INTENT DETECTION AND SLOT FILLING USING CONVOLUTIONAL NEURAL NETWORKS. Puyang Xu, Ruhi Sarikaya. Microsoft Corporation

JOINT INTENT DETECTION AND SLOT FILLING USING CONVOLUTIONAL NEURAL NETWORKS. Puyang Xu, Ruhi Sarikaya. Microsoft Corporation JOINT INTENT DETECTION AND SLOT FILLING USING CONVOLUTIONAL NEURAL NETWORKS Puyang Xu, Ruhi Sarikaya Microsoft Corporation ABSTRACT We describe a joint model for intent detection and slot filling based

More information

Combine the PA Algorithm with a Proximal Classifier

Combine the PA Algorithm with a Proximal Classifier Combine the Passive and Aggressive Algorithm with a Proximal Classifier Yuh-Jye Lee Joint work with Y.-C. Tseng Dept. of Computer Science & Information Engineering TaiwanTech. Dept. of Statistics@NCKU

More information

Advanced Introduction to Machine Learning, CMU-10715

Advanced Introduction to Machine Learning, CMU-10715 Advanced Introduction to Machine Learning, CMU-10715 Deep Learning Barnabás Póczos, Sept 17 Credits Many of the pictures, results, and other materials are taken from: Ruslan Salakhutdinov Joshua Bengio

More information

Machine Learning Basics. Sargur N. Srihari

Machine Learning Basics. Sargur N. Srihari Machine Learning Basics Sargur N. srihari@cedar.buffalo.edu 1 Overview Deep learning is a specific type of ML Necessary to have a solid understanding of the basic principles of ML 2 Topics Stochastic Gradient

More information

GraphGAN: Graph Representation Learning with Generative Adversarial Nets

GraphGAN: Graph Representation Learning with Generative Adversarial Nets The 32 nd AAAI Conference on Artificial Intelligence (AAAI 2018) New Orleans, Louisiana, USA GraphGAN: Graph Representation Learning with Generative Adversarial Nets Hongwei Wang 1,2, Jia Wang 3, Jialin

More information

CS6375: Machine Learning Gautam Kunapuli. Mid-Term Review

CS6375: Machine Learning Gautam Kunapuli. Mid-Term Review Gautam Kunapuli Machine Learning Data is identically and independently distributed Goal is to learn a function that maps to Data is generated using an unknown function Learn a hypothesis that minimizes

More information

Machine Learning Lecture 3

Machine Learning Lecture 3 Many slides adapted from B. Schiele Machine Learning Lecture 3 Probability Density Estimation II 26.04.2016 Bastian Leibe RWTH Aachen http://www.vision.rwth-aachen.de leibe@vision.rwth-aachen.de Course

More information

Neural Network Optimization and Tuning / Spring 2018 / Recitation 3

Neural Network Optimization and Tuning / Spring 2018 / Recitation 3 Neural Network Optimization and Tuning 11-785 / Spring 2018 / Recitation 3 1 Logistics You will work through a Jupyter notebook that contains sample and starter code with explanations and comments throughout.

More information

Introduction to Deep Learning

Introduction to Deep Learning ENEE698A : Machine Learning Seminar Introduction to Deep Learning Raviteja Vemulapalli Image credit: [LeCun 1998] Resources Unsupervised feature learning and deep learning (UFLDL) tutorial (http://ufldl.stanford.edu/wiki/index.php/ufldl_tutorial)

More information

Deep Learning for Program Analysis. Lili Mou January, 2016

Deep Learning for Program Analysis. Lili Mou January, 2016 Deep Learning for Program Analysis Lili Mou January, 2016 Outline Introduction Background Deep Neural Networks Real-Valued Representation Learning Our Models Building Program Vector Representations for

More information

Deep Tracking: Biologically Inspired Tracking with Deep Convolutional Networks

Deep Tracking: Biologically Inspired Tracking with Deep Convolutional Networks Deep Tracking: Biologically Inspired Tracking with Deep Convolutional Networks Si Chen The George Washington University sichen@gwmail.gwu.edu Meera Hahn Emory University mhahn7@emory.edu Mentor: Afshin

More information

Hidden Markov Models. Gabriela Tavares and Juri Minxha Mentor: Taehwan Kim CS159 04/25/2017

Hidden Markov Models. Gabriela Tavares and Juri Minxha Mentor: Taehwan Kim CS159 04/25/2017 Hidden Markov Models Gabriela Tavares and Juri Minxha Mentor: Taehwan Kim CS159 04/25/2017 1 Outline 1. 2. 3. 4. Brief review of HMMs Hidden Markov Support Vector Machines Large Margin Hidden Markov Models

More information

Machine Learning Lecture 3

Machine Learning Lecture 3 Course Outline Machine Learning Lecture 3 Fundamentals (2 weeks) Bayes Decision Theory Probability Density Estimation Probability Density Estimation II 26.04.206 Discriminative Approaches (5 weeks) Linear

More information

Perceptron Introduction to Machine Learning. Matt Gormley Lecture 5 Jan. 31, 2018

Perceptron Introduction to Machine Learning. Matt Gormley Lecture 5 Jan. 31, 2018 10-601 Introduction to Machine Learning Machine Learning Department School of Computer Science Carnegie Mellon University Perceptron Matt Gormley Lecture 5 Jan. 31, 2018 1 Q&A Q: We pick the best hyperparameters

More information

Expectation Maximization (EM) and Gaussian Mixture Models

Expectation Maximization (EM) and Gaussian Mixture Models Expectation Maximization (EM) and Gaussian Mixture Models Reference: The Elements of Statistical Learning, by T. Hastie, R. Tibshirani, J. Friedman, Springer 1 2 3 4 5 6 7 8 Unsupervised Learning Motivation

More information

CAMCOS Report Day. December 9 th, 2015 San Jose State University Project Theme: Classification

CAMCOS Report Day. December 9 th, 2015 San Jose State University Project Theme: Classification CAMCOS Report Day December 9 th, 2015 San Jose State University Project Theme: Classification On Classification: An Empirical Study of Existing Algorithms based on two Kaggle Competitions Team 1 Team 2

More information

Large-Scale Lasso and Elastic-Net Regularized Generalized Linear Models

Large-Scale Lasso and Elastic-Net Regularized Generalized Linear Models Large-Scale Lasso and Elastic-Net Regularized Generalized Linear Models DB Tsai Steven Hillion Outline Introduction Linear / Nonlinear Classification Feature Engineering - Polynomial Expansion Big-data

More information

CSE 573: Artificial Intelligence Autumn 2010

CSE 573: Artificial Intelligence Autumn 2010 CSE 573: Artificial Intelligence Autumn 2010 Lecture 16: Machine Learning Topics 12/7/2010 Luke Zettlemoyer Most slides over the course adapted from Dan Klein. 1 Announcements Syllabus revised Machine

More information

Machine Learning. Chao Lan

Machine Learning. Chao Lan Machine Learning Chao Lan Machine Learning Prediction Models Regression Model - linear regression (least square, ridge regression, Lasso) Classification Model - naive Bayes, logistic regression, Gaussian

More information

Computer vision: models, learning and inference. Chapter 10 Graphical Models

Computer vision: models, learning and inference. Chapter 10 Graphical Models Computer vision: models, learning and inference Chapter 10 Graphical Models Independence Two variables x 1 and x 2 are independent if their joint probability distribution factorizes as Pr(x 1, x 2 )=Pr(x

More information

Autoencoder. Representation learning (related to dictionary learning) Both the input and the output are x

Autoencoder. Representation learning (related to dictionary learning) Both the input and the output are x Deep Learning 4 Autoencoder, Attention (spatial transformer), Multi-modal learning, Neural Turing Machine, Memory Networks, Generative Adversarial Net Jian Li IIIS, Tsinghua Autoencoder Autoencoder Unsupervised

More information

CS489/698: Intro to ML

CS489/698: Intro to ML CS489/698: Intro to ML Lecture 14: Training of Deep NNs Instructor: Sun Sun 1 Outline Activation functions Regularization Gradient-based optimization 2 Examples of activation functions 3 5/28/18 Sun Sun

More information

Machine Learning: Think Big and Parallel

Machine Learning: Think Big and Parallel Day 1 Inderjit S. Dhillon Dept of Computer Science UT Austin CS395T: Topics in Multicore Programming Oct 1, 2013 Outline Scikit-learn: Machine Learning in Python Supervised Learning day1 Regression: Least

More information

Cheng Soon Ong & Christian Walder. Canberra February June 2018

Cheng Soon Ong & Christian Walder. Canberra February June 2018 Cheng Soon Ong & Christian Walder Research Group and College of Engineering and Computer Science Canberra February June 2018 Outlines Overview Introduction Linear Algebra Probability Linear Regression

More information

Energy Based Models, Restricted Boltzmann Machines and Deep Networks. Jesse Eickholt

Energy Based Models, Restricted Boltzmann Machines and Deep Networks. Jesse Eickholt Energy Based Models, Restricted Boltzmann Machines and Deep Networks Jesse Eickholt ???? Who s heard of Energy Based Models (EBMs) Restricted Boltzmann Machines (RBMs) Deep Belief Networks Auto-encoders

More information

Machine Learning. The Breadth of ML Neural Networks & Deep Learning. Marc Toussaint. Duy Nguyen-Tuong. University of Stuttgart

Machine Learning. The Breadth of ML Neural Networks & Deep Learning. Marc Toussaint. Duy Nguyen-Tuong. University of Stuttgart Machine Learning The Breadth of ML Neural Networks & Deep Learning Marc Toussaint University of Stuttgart Duy Nguyen-Tuong Bosch Center for Artificial Intelligence Summer 2017 Neural Networks Consider

More information

Contents. Preface to the Second Edition

Contents. Preface to the Second Edition Preface to the Second Edition v 1 Introduction 1 1.1 What Is Data Mining?....................... 4 1.2 Motivating Challenges....................... 5 1.3 The Origins of Data Mining....................

More information

Network embedding. Cheng Zheng

Network embedding. Cheng Zheng Network embedding Cheng Zheng Outline Problem definition Factorization based algorithms --- Laplacian Eigenmaps(NIPS, 2001) Random walk based algorithms ---DeepWalk(KDD, 2014), node2vec(kdd, 2016) Deep

More information

COMP 551 Applied Machine Learning Lecture 13: Unsupervised learning

COMP 551 Applied Machine Learning Lecture 13: Unsupervised learning COMP 551 Applied Machine Learning Lecture 13: Unsupervised learning Associate Instructor: Herke van Hoof (herke.vanhoof@mail.mcgill.ca) Slides mostly by: (jpineau@cs.mcgill.ca) Class web page: www.cs.mcgill.ca/~jpineau/comp551

More information

08 An Introduction to Dense Continuous Robotic Mapping

08 An Introduction to Dense Continuous Robotic Mapping NAVARCH/EECS 568, ROB 530 - Winter 2018 08 An Introduction to Dense Continuous Robotic Mapping Maani Ghaffari March 14, 2018 Previously: Occupancy Grid Maps Pose SLAM graph and its associated dense occupancy

More information

All lecture slides will be available at CSC2515_Winter15.html

All lecture slides will be available at  CSC2515_Winter15.html CSC2515 Fall 2015 Introduc3on to Machine Learning Lecture 9: Support Vector Machines All lecture slides will be available at http://www.cs.toronto.edu/~urtasun/courses/csc2515/ CSC2515_Winter15.html Many

More information

Partitioning Data. IRDS: Evaluation, Debugging, and Diagnostics. Cross-Validation. Cross-Validation for parameter tuning

Partitioning Data. IRDS: Evaluation, Debugging, and Diagnostics. Cross-Validation. Cross-Validation for parameter tuning Partitioning Data IRDS: Evaluation, Debugging, and Diagnostics Charles Sutton University of Edinburgh Training Validation Test Training : Running learning algorithms Validation : Tuning parameters of learning

More information

Deep Learning & Neural Networks

Deep Learning & Neural Networks Deep Learning & Neural Networks Machine Learning CSE4546 Sham Kakade University of Washington November 29, 2016 Sham Kakade 1 Announcements: HW4 posted Poster Session Thurs, Dec 8 Today: Review: EM Neural

More information

Deep Boosting. Joint work with Corinna Cortes (Google Research) Vitaly Kuznetsov (Courant Institute) Umar Syed (Google Research)

Deep Boosting. Joint work with Corinna Cortes (Google Research) Vitaly Kuznetsov (Courant Institute) Umar Syed (Google Research) Deep Boosting Joint work with Corinna Cortes (Google Research) Vitaly Kuznetsov (Courant Institute) Umar Syed (Google Research) MEHRYAR MOHRI MOHRI@ COURANT INSTITUTE & GOOGLE RESEARCH. Deep Boosting Essence

More information

Efficient Algorithms may not be those we think

Efficient Algorithms may not be those we think Efficient Algorithms may not be those we think Yann LeCun, Computational and Biological Learning Lab The Courant Institute of Mathematical Sciences New York University http://yann.lecun.com http://www.cs.nyu.edu/~yann

More information

All You Want To Know About CNNs. Yukun Zhu

All You Want To Know About CNNs. Yukun Zhu All You Want To Know About CNNs Yukun Zhu Deep Learning Deep Learning Image from http://imgur.com/ Deep Learning Image from http://imgur.com/ Deep Learning Image from http://imgur.com/ Deep Learning Image

More information

Global Optimality in Neural Network Training

Global Optimality in Neural Network Training Global Optimality in Neural Network Training Benjamin D. Haeffele and René Vidal Johns Hopkins University, Center for Imaging Science. Baltimore, USA Questions in Deep Learning Architecture Design Optimization

More information

Clustering K-means. Machine Learning CSEP546 Carlos Guestrin University of Washington February 18, Carlos Guestrin

Clustering K-means. Machine Learning CSEP546 Carlos Guestrin University of Washington February 18, Carlos Guestrin Clustering K-means Machine Learning CSEP546 Carlos Guestrin University of Washington February 18, 2014 Carlos Guestrin 2005-2014 1 Clustering images Set of Images [Goldberger et al.] Carlos Guestrin 2005-2014

More information

Deep Learning Workshop. Nov. 20, 2015 Andrew Fishberg, Rowan Zellers

Deep Learning Workshop. Nov. 20, 2015 Andrew Fishberg, Rowan Zellers Deep Learning Workshop Nov. 20, 2015 Andrew Fishberg, Rowan Zellers Why deep learning? The ImageNet Challenge Goal: image classification with 1000 categories Top 5 error rate of 15%. Krizhevsky, Alex,

More information

Theoretical Concepts of Machine Learning

Theoretical Concepts of Machine Learning Theoretical Concepts of Machine Learning Part 2 Institute of Bioinformatics Johannes Kepler University, Linz, Austria Outline 1 Introduction 2 Generalization Error 3 Maximum Likelihood 4 Noise Models 5

More information

On Information-Maximization Clustering: Tuning Parameter Selection and Analytic Solution

On Information-Maximization Clustering: Tuning Parameter Selection and Analytic Solution ICML2011 Jun. 28-Jul. 2, 2011 On Information-Maximization Clustering: Tuning Parameter Selection and Analytic Solution Masashi Sugiyama, Makoto Yamada, Manabu Kimura, and Hirotaka Hachiya Department of

More information

Stacked Denoising Autoencoders for Face Pose Normalization

Stacked Denoising Autoencoders for Face Pose Normalization Stacked Denoising Autoencoders for Face Pose Normalization Yoonseop Kang 1, Kang-Tae Lee 2,JihyunEun 2, Sung Eun Park 2 and Seungjin Choi 1 1 Department of Computer Science and Engineering Pohang University

More information

Neural Networks and Deep Learning

Neural Networks and Deep Learning Neural Networks and Deep Learning Example Learning Problem Example Learning Problem Celebrity Faces in the Wild Machine Learning Pipeline Raw data Feature extract. Feature computation Inference: prediction,

More information

CS570: Introduction to Data Mining

CS570: Introduction to Data Mining CS570: Introduction to Data Mining Classification Advanced Reading: Chapter 8 & 9 Han, Chapters 4 & 5 Tan Anca Doloc-Mihu, Ph.D. Slides courtesy of Li Xiong, Ph.D., 2011 Han, Kamber & Pei. Data Mining.

More information

International Journal of Computer Engineering and Applications, Volume XII, Special Issue, September 18,

International Journal of Computer Engineering and Applications, Volume XII, Special Issue, September 18, REAL-TIME OBJECT DETECTION WITH CONVOLUTION NEURAL NETWORK USING KERAS Asmita Goswami [1], Lokesh Soni [2 ] Department of Information Technology [1] Jaipur Engineering College and Research Center Jaipur[2]

More information

Classification by Support Vector Machines

Classification by Support Vector Machines Classification by Support Vector Machines Florian Markowetz Max-Planck-Institute for Molecular Genetics Computational Molecular Biology Berlin Practical DNA Microarray Analysis 2003 1 Overview I II III

More information

Autoencoders, denoising autoencoders, and learning deep networks

Autoencoders, denoising autoencoders, and learning deep networks 4 th CiFAR Summer School on Learning and Vision in Biology and Engineering Toronto, August 5-9 2008 Autoencoders, denoising autoencoders, and learning deep networks Part II joint work with Hugo Larochelle,

More information