Efficient Complex Output Prediction

Size: px
Start display at page:

Download "Efficient Complex Output Prediction"

Transcription

1 Efficient Complex Output Prediction Florence d Alché-Buc Joint work with Romain Brault, Alex Lambert, Maxime Sangnier October 12, 2017 LTCI, Télécom ParisTech, Institut-Mines Télécom, Université Paris-Saclay

2 Outline Motivation and Goals Operator-valued Kernel Regression Scaling up Operator-valued Kernel Regression Conclusion 1

3 Classic Regression Using training data {(x i, y i ), i = 1,... N}, build a scalar-valued function f that predicts an output y R, given some input x X : Complex output regression: when Y = R p or a structured objects set or a functional space 2

4 Multiple Output Regression When Y = R p Image understanding : predict the name of an object in an image X : image representation space, Y = R p : semantic space Joint Quantile Regression Y = R p as a multitask learning for the values of the wished quantiles: τ 1,..., τ p (Sangnier et al. 2016) 3

5 Multiple Output Regression When Y is set of structured objects Identification of metabolites from mass spectra X : mass spectra space, Y= set of metabolites When Y = F a space of functions Functional quantile regression X = R d and Y = H a Reproducing Kernel Hilbert Space (Brault 2017) 4

6 Learning functions with values in a Hilbert space Y Operator-valued kernels, vector-valued Reproducing Kernel Hilbert Spaces Nonparametric learning Various loss functions for data-fitting Various kinds of regularization With theoretical guarantees both in terms of statistics and optimization that also leads to efficient learning algorithms 5

7 Outline Motivation and Goals Operator-valued Kernel Regression Scaling up Operator-valued Kernel Regression Conclusion 6

8 Operator-valued Kernels Natural extension of scalar kernels for vector-valued functions Allows coupling between outputs Let X some input space and Y a Hilbert Space Scalar Operator-valued Domain k(x, z) R K(x, z) L(Y) Symmetry k(x, z) = k(z, x) K(x, z) = K(z, x) N PD c i, c j R, i,j=1 c ic j k(x i, x j ) 0 c i, c j R p N, i,j=1 < c i, K(x i, x j )c j > Y 0 The simplest operator-valued kernel, the decomposable one: K(x, x ) = k(x, x )B, where B a positive semi-definite p p matrix; k is scalar-valued kernel on X. B = I : recovers the case of p independent scalar-valued kernel machines. 7

9 Vector-valued RKHS Given an OVK K, unique vector-valued RKHS (H K ), Feature maps: K(x, z) = Φ(x) Φ(z), Representer theorems. Representer theorem (Micchelli and Pontil, 2005) Given a training set {(x 1, y 1 ),..., (x N, y N )} X Y, the minimizer N ˆf = arg minf f 2 H K + λ N l(y i, f(x i )) admits an expansion of the form: where c i Y. i=1 ˆf( ) N = K(, x i )c i, i=1 8

10 Regression in Vector-valued RKHS: a few examples Image understanding : predict object name in an image Surrogate loss l Fisher (y, f(x)) = θ ln p θ (y) f(x) 2 + pre-image (Djerrab et al. 2017) Decomposable kernels Very good results on Few-shot-learning (Caltech101) 9

11 Regression in Vector-valued RKHS: a few examples Image understanding : predict object name in an image Surrogate loss l Fisher (y, f(x)) = θ ln p θ (y) f(x) 2 + pre-image (Djerrab et al. 2017) Decomposable kernels Very good results on Few-shot-learning (Caltech101) Sparse modeling of time series Loss: ϵ-sensitive loss and transformable kernels (Lim et al. 2013, 2015; Sangnier et al. 2016) Application : modeling climate data (Lim et al. 2015) 9

12 Joint Quantile Regression as multitask learning Loss: l pinball (y, f(x)) = l τ (y.1 f(x) b), y R but f(x) R p Decomposable matrix parameterized with the values of the wished quantiles: τ 1,..., τ p (Sangnier et al. 2016) Pinball loss: l τ (r) = { p j=1 τ j r j if r j 0, (τ j 1)r j if r j < 0. 10

13 Outline Motivation and Goals Operator-valued Kernel Regression Scaling up Operator-valued Kernel Regression Conclusion 11

14 Scalability of Regression in vector-valued RKHS Focus on Kernel ridge regression for Y = R p : Prediction in linear time w.r.t data O(Np 2 ), Naive learning (closed form) in O(N 3 p 3 ) How to make the method scalable? Find a matrix-valued feature map, ϕ, such that K(x, z) K(x, z) = Φ(x) Φ(z), (1) In order to work with the following linear model f(x) = Φ(x) θ (2) where θ R D. 12

15 Toward spectral approximation of OVK Theorem (Bochner for OVK (Carmeli et al. 2010)) Let K: R d R d L(Y) be a translation invariant positive definite continuous OVK. There exists a unique non-negative Borel operator-valued measure Q such that x, z R d R d K(x, z) = cos ( x z, ω )dq(ω) R d 13

16 Toward spectral approximation of OVK Theorem (Bochner for OVK (Carmeli et al. 2010)) Let K: R d R d L(Y) be a translation invariant positive definite continuous OVK. There exists a unique non-negative Borel operator-valued measure Q such that x, z R d R d K(x, z) = cos ( x z, ω )dq(ω) R d Find B : R d L(U; Y), µ scalar positive measure, such that dq(ω) = B(ω)B(ω) dµ(ω) 13

17 Operator Random Fourier Features (ORFF) Assume µ is a probability distribution Then given (ω j ) D j=1 µd i.i.d construct (Brault et al. 2016) Φ: X L(Y, U 2D ) ( ) cos( x, ωj )B(ω j ) x 1 D D j=1 sin( x, ω j )B(ω j ) Φ approximated feature map for kernel K. x, z R d R d, Φ(x) Φ(z) = 1 D D cos ( x z, ω j )B(ω j )B(ω j ) K(x, z) D j=1 where the convergence holds µ-almost everywhere in the weak sense. 14

18 Application to Functional Quantile Regression Toy dataset: N = 1000 points, D = 100(ORFF on input kernel) and D = 100(RFF on output kernel) Matches performance obtained by multi-task learning (Sangnier et al. 2016), but faster, and with an access to all quantiles levels. 15

19 Outline Motivation and Goals Operator-valued Kernel Regression Scaling up Operator-valued Kernel Regression Conclusion 16

20 Conclusion Operator-valued Kernel Regression: extends kernel methods to more involved prediction problems Versatile framework : losses and kernels Scalability obtained with Random Fourier Feature techniques Theoretical guarantees on approximation 17

21 Perspectives Theoretical properties of learning with ORFF Stacking ORFF / links with Deep Learning Towards Hybrid Architectures (Mairal 2016) Image/text understanding (combining Deep Neural architectures and ORFF) Anomaly detection (extending one-class SVM) Spatio-temporal data : climatics, epidemics data 18

22 Thank you for your attention Our contributions C. Brouard, F. d Alché-Buc, M.Szafranski: Semi-supervised Penalized Output Kernel Regression for Link Prediction. ICML 2011: N. Lim, Y. Senbabaoglu, G.Michailidis, F. d Alché-Buc: OKVAR-Boost: a novel boosting algorithm to infer nonlinear dynamics and interactions in gene regulatory networks. Bioinformatics 29(11): (2013) N. Lim, F. d Alché-Buc, C. Auliac, G. Michailidis, Operator-valued Kernel based Vector Autoregressive Models for Network Inference, Machine Learning Journal, April C. Brouard, M. Szafranski, F. d Alché-Buc,, Input Output Kernel Regression for supervised and semi-supervised structured output learning, JMLR, 2016 C. Brouard, H. Shen, K. Dührkop, F. d Alché-Buc, S. Böcker, J. Rousu: Fast metabolite identification with Input Output Kernel Regression. Bioinformatics 32(12): (2016) M. Sangnier, O. Fercoq, F. d Alché-Buc, Joint quantile regression in vector-valued RKHSs, NIPS 2016: , (2016) R. Brault, M. Heinonen, F. d Alché-Buc, Random Fourier Features for Operator-valued Kernels, ACML 2016, (2016) M. Djerrab, A. Garcia, M. Sangnier, F. d Alché-Buc, Output Fisher Embedding Regression, Machine Learning Journal (in revision), (2017) M. Sangnier, O. Fercoq, F. d Alché-Buc, Data sparse nonparametric regression with ϵ-insensitive losses, ACML 2017, (2017) 19

23 Collaborations are welcome One-year postdoc position (from January) Master internship positions (April-september) co-supervising PhD thesis Contact: 20

Scaling up Vector Autoregressive Models with Operator Random Fourier Features

Scaling up Vector Autoregressive Models with Operator Random Fourier Features Scaling up Vector Autoregressive Models with Operator Random Fourier Features Romain Brault, Néhémy Lim, Florence d Alché-Buc August, 6 Abstract A nonparametric approach to Vector Autoregressive Modeling

More information

Kernel Learning via Random Fourier Representations

Kernel Learning via Random Fourier Representations Kernel Learning via Random Fourier Representations L. Law, M. Mider, X. Miscouridou, S. Ip, A. Wang Module 5: Machine Learning L. Law, M. Mider, X. Miscouridou, S. Ip, A. Wang Kernel Learning via Random

More information

Approximate Kernel PCA with Random Features

Approximate Kernel PCA with Random Features Approximate Kernel PCA with Random Features (Computational vs. Statistical Tradeoff) Bharath K. Sriperumbudur Department of Statistics, Pennsylvania State University Journées de Statistique Paris May 28,

More information

Data sparse nonparametric regression with ɛ-insensitive losses

Data sparse nonparametric regression with ɛ-insensitive losses Proceedings of Machine Learning Research 77:92 207, 207 ACML 207 Data sparse nonparametric regression with ɛ-insensitive losses Maxime Sangnier maxime.sangnier@upmc.fr Sorbonne Universités, UPMC Univ Paris

More information

Approximate Kernel Methods

Approximate Kernel Methods Lecture 3 Approximate Kernel Methods Bharath K. Sriperumbudur Department of Statistics, Pennsylvania State University Machine Learning Summer School Tübingen, 207 Outline Motivating example Ridge regression

More information

Learning gradients: prescriptive models

Learning gradients: prescriptive models Department of Statistical Science Institute for Genome Sciences & Policy Department of Computer Science Duke University May 11, 2007 Relevant papers Learning Coordinate Covariances via Gradients. Sayan

More information

Mathematical Methods for Data Analysis

Mathematical Methods for Data Analysis Mathematical Methods for Data Analysis Massimiliano Pontil Istituto Italiano di Tecnologia and Department of Computer Science University College London Massimiliano Pontil Mathematical Methods for Data

More information

Diffeomorphic Warping. Ben Recht August 17, 2006 Joint work with Ali Rahimi (Intel)

Diffeomorphic Warping. Ben Recht August 17, 2006 Joint work with Ali Rahimi (Intel) Diffeomorphic Warping Ben Recht August 17, 2006 Joint work with Ali Rahimi (Intel) What Manifold Learning Isn t Common features of Manifold Learning Algorithms: 1-1 charting Dense sampling Geometric Assumptions

More information

MATH 829: Introduction to Data Mining and Analysis Support vector machines and kernels

MATH 829: Introduction to Data Mining and Analysis Support vector machines and kernels 1/12 MATH 829: Introduction to Data Mining and Analysis Support vector machines and kernels Dominique Guillot Departments of Mathematical Sciences University of Delaware March 14, 2016 Separating sets:

More information

10-701/ Recitation : Kernels

10-701/ Recitation : Kernels 10-701/15-781 Recitation : Kernels Manojit Nandi February 27, 2014 Outline Mathematical Theory Banach Space and Hilbert Spaces Kernels Commonly Used Kernels Kernel Theory One Weird Kernel Trick Representer

More information

Kernels A Machine Learning Overview

Kernels A Machine Learning Overview Kernels A Machine Learning Overview S.V.N. Vishy Vishwanathan vishy@axiom.anu.edu.au National ICT of Australia and Australian National University Thanks to Alex Smola, Stéphane Canu, Mike Jordan and Peter

More information

Bits of Machine Learning Part 1: Supervised Learning

Bits of Machine Learning Part 1: Supervised Learning Bits of Machine Learning Part 1: Supervised Learning Alexandre Proutiere and Vahan Petrosyan KTH (The Royal Institute of Technology) Outline of the Course 1. Supervised Learning Regression and Classification

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

Fast metabolite identification with Input Output Kernel Regression

Fast metabolite identification with Input Output Kernel Regression Bioinformatics doi.10.1093/bioinformatics/xxxxxx Advance Access Publication Date: Day Month Year Manuscript Category Fast metabolite identification with Input Output Kernel Regression Céline Brouard 1,2,,

More information

Kernels for Multi task Learning

Kernels for Multi task Learning Kernels for Multi task Learning Charles A Micchelli Department of Mathematics and Statistics State University of New York, The University at Albany 1400 Washington Avenue, Albany, NY, 12222, USA Massimiliano

More information

TUM 2016 Class 3 Large scale learning by regularization

TUM 2016 Class 3 Large scale learning by regularization TUM 2016 Class 3 Large scale learning by regularization Lorenzo Rosasco UNIGE-MIT-IIT July 25, 2016 Learning problem Solve min w E(w), E(w) = dρ(x, y)l(w x, y) given (x 1, y 1 ),..., (x n, y n ) Beyond

More information

Advanced Introduction to Machine Learning

Advanced Introduction to Machine Learning 10-715 Advanced Introduction to Machine Learning Homework Due Oct 15, 10.30 am Rules Please follow these guidelines. Failure to do so, will result in loss of credit. 1. Homework is due on the due date

More information

CIS 520: Machine Learning Oct 09, Kernel Methods

CIS 520: Machine Learning Oct 09, Kernel Methods CIS 520: Machine Learning Oct 09, 207 Kernel Methods Lecturer: Shivani Agarwal Disclaimer: These notes are designed to be a supplement to the lecture They may or may not cover all the material discussed

More information

Basis Expansion and Nonlinear SVM. Kai Yu

Basis Expansion and Nonlinear SVM. Kai Yu Basis Expansion and Nonlinear SVM Kai Yu Linear Classifiers f(x) =w > x + b z(x) = sign(f(x)) Help to learn more general cases, e.g., nonlinear models 8/7/12 2 Nonlinear Classifiers via Basis Expansion

More information

Kernels for Automatic Pattern Discovery and Extrapolation

Kernels for Automatic Pattern Discovery and Extrapolation Kernels for Automatic Pattern Discovery and Extrapolation Andrew Gordon Wilson agw38@cam.ac.uk mlg.eng.cam.ac.uk/andrew University of Cambridge Joint work with Ryan Adams (Harvard) 1 / 21 Pattern Recognition

More information

Robust Support Vector Machines for Probability Distributions

Robust Support Vector Machines for Probability Distributions Robust Support Vector Machines for Probability Distributions Andreas Christmann joint work with Ingo Steinwart (Los Alamos National Lab) ICORS 2008, Antalya, Turkey, September 8-12, 2008 Andreas Christmann,

More information

Stochastic optimization in Hilbert spaces

Stochastic optimization in Hilbert spaces Stochastic optimization in Hilbert spaces Aymeric Dieuleveut Aymeric Dieuleveut Stochastic optimization Hilbert spaces 1 / 48 Outline Learning vs Statistics Aymeric Dieuleveut Stochastic optimization Hilbert

More information

Operator-valued kernel-based vector autoregressive models for network inference

Operator-valued kernel-based vector autoregressive models for network inference Mach Learn (2015) 99:489 513 DOI 10.1007/s10994-014-5479-3 Operator-valued kernel-based vector autoregressive models for network inference Néhémy Lim Florence d Alché-Buc Cédric Auliac George Michailidis

More information

Support Vector Machines for Classification: A Statistical Portrait

Support Vector Machines for Classification: A Statistical Portrait Support Vector Machines for Classification: A Statistical Portrait Yoonkyung Lee Department of Statistics The Ohio State University May 27, 2011 The Spring Conference of Korean Statistical Society KAIST,

More information

Simple Optimization, Bigger Models, and Faster Learning. Niao He

Simple Optimization, Bigger Models, and Faster Learning. Niao He Simple Optimization, Bigger Models, and Faster Learning Niao He Big Data Symposium, UIUC, 2016 Big Data, Big Picture Niao He (UIUC) 2/26 Big Data, Big Picture Niao He (UIUC) 3/26 Big Data, Big Picture

More information

Statistical learning theory, Support vector machines, and Bioinformatics

Statistical learning theory, Support vector machines, and Bioinformatics 1 Statistical learning theory, Support vector machines, and Bioinformatics Jean-Philippe.Vert@mines.org Ecole des Mines de Paris Computational Biology group ENS Paris, november 25, 2003. 2 Overview 1.

More information

Stein Variational Gradient Descent: A General Purpose Bayesian Inference Algorithm

Stein Variational Gradient Descent: A General Purpose Bayesian Inference Algorithm Stein Variational Gradient Descent: A General Purpose Bayesian Inference Algorithm Qiang Liu and Dilin Wang NIPS 2016 Discussion by Yunchen Pu March 17, 2017 March 17, 2017 1 / 8 Introduction Let x R d

More information

MTTTS16 Learning from Multiple Sources

MTTTS16 Learning from Multiple Sources MTTTS16 Learning from Multiple Sources 5 ECTS credits Autumn 2018, University of Tampere Lecturer: Jaakko Peltonen Lecture 6: Multitask learning with kernel methods and nonparametric models On this lecture:

More information

Reproducing Kernel Hilbert Spaces

Reproducing Kernel Hilbert Spaces Reproducing Kernel Hilbert Spaces Lorenzo Rosasco 9.520 Class 03 February 11, 2009 About this class Goal To introduce a particularly useful family of hypothesis spaces called Reproducing Kernel Hilbert

More information

Kernel Methods in Machine Learning

Kernel Methods in Machine Learning Kernel Methods in Machine Learning Autumn 2015 Lecture 1: Introduction Juho Rousu ICS-E4030 Kernel Methods in Machine Learning 9. September, 2015 uho Rousu (ICS-E4030 Kernel Methods in Machine Learning)

More information

A Magiv CV Theory for Large-Margin Classifiers

A Magiv CV Theory for Large-Margin Classifiers A Magiv CV Theory for Large-Margin Classifiers Hui Zou School of Statistics, University of Minnesota June 30, 2018 Joint work with Boxiang Wang Outline 1 Background 2 Magic CV formula 3 Magic support vector

More information

Reproducing Kernel Hilbert Spaces

Reproducing Kernel Hilbert Spaces Reproducing Kernel Hilbert Spaces Lorenzo Rosasco 9.520 Class 03 February 12, 2007 About this class Goal To introduce a particularly useful family of hypothesis spaces called Reproducing Kernel Hilbert

More information

Kernel-Based Contrast Functions for Sufficient Dimension Reduction

Kernel-Based Contrast Functions for Sufficient Dimension Reduction Kernel-Based Contrast Functions for Sufficient Dimension Reduction Michael I. Jordan Departments of Statistics and EECS University of California, Berkeley Joint work with Kenji Fukumizu and Francis Bach

More information

An Adaptive Test of Independence with Analytic Kernel Embeddings

An Adaptive Test of Independence with Analytic Kernel Embeddings An Adaptive Test of Independence with Analytic Kernel Embeddings Wittawat Jitkrittum 1 Zoltán Szabó 2 Arthur Gretton 1 1 Gatsby Unit, University College London 2 CMAP, École Polytechnique ICML 2017, Sydney

More information

Weak Signals: machine-learning meets extreme value theory

Weak Signals: machine-learning meets extreme value theory Weak Signals: machine-learning meets extreme value theory Stephan Clémençon Télécom ParisTech, LTCI, Université Paris Saclay machinelearningforbigdata.telecom-paristech.fr 2017-10-12, Workshop Big Data

More information

Dimensionality Reduction and Principle Components Analysis

Dimensionality Reduction and Principle Components Analysis Dimensionality Reduction and Principle Components Analysis 1 Outline What is dimensionality reduction? Principle Components Analysis (PCA) Example (Bishop, ch 12) PCA vs linear regression PCA as a mixture

More information

Linear Regression 1 / 25. Karl Stratos. June 18, 2018

Linear Regression 1 / 25. Karl Stratos. June 18, 2018 Linear Regression Karl Stratos June 18, 2018 1 / 25 The Regression Problem Problem. Find a desired input-output mapping f : X R where the output is a real value. x = = y = 0.1 How much should I turn my

More information

Memory Efficient Kernel Approximation

Memory Efficient Kernel Approximation Si Si Department of Computer Science University of Texas at Austin ICML Beijing, China June 23, 2014 Joint work with Cho-Jui Hsieh and Inderjit S. Dhillon Outline Background Motivation Low-Rank vs. Block

More information

RKHS, Mercer s theorem, Unbounded domains, Frames and Wavelets Class 22, 2004 Tomaso Poggio and Sayan Mukherjee

RKHS, Mercer s theorem, Unbounded domains, Frames and Wavelets Class 22, 2004 Tomaso Poggio and Sayan Mukherjee RKHS, Mercer s theorem, Unbounded domains, Frames and Wavelets 9.520 Class 22, 2004 Tomaso Poggio and Sayan Mukherjee About this class Goal To introduce an alternate perspective of RKHS via integral operators

More information

Convergence Rates of Kernel Quadrature Rules

Convergence Rates of Kernel Quadrature Rules Convergence Rates of Kernel Quadrature Rules Francis Bach INRIA - Ecole Normale Supérieure, Paris, France ÉCOLE NORMALE SUPÉRIEURE NIPS workshop on probabilistic integration - Dec. 2015 Outline Introduction

More information

Statistical learning on graphs

Statistical learning on graphs Statistical learning on graphs Jean-Philippe Vert Jean-Philippe.Vert@ensmp.fr ParisTech, Ecole des Mines de Paris Institut Curie INSERM U900 Seminar of probabilities, Institut Joseph Fourier, Grenoble,

More information

ECS289: Scalable Machine Learning

ECS289: Scalable Machine Learning ECS289: Scalable Machine Learning Cho-Jui Hsieh UC Davis Oct 18, 2016 Outline One versus all/one versus one Ranking loss for multiclass/multilabel classification Scaling to millions of labels Multiclass

More information

Indirect Rule Learning: Support Vector Machines. Donglin Zeng, Department of Biostatistics, University of North Carolina

Indirect Rule Learning: Support Vector Machines. Donglin Zeng, Department of Biostatistics, University of North Carolina Indirect Rule Learning: Support Vector Machines Indirect learning: loss optimization It doesn t estimate the prediction rule f (x) directly, since most loss functions do not have explicit optimizers. Indirection

More information

Beyond the Point Cloud: From Transductive to Semi-Supervised Learning

Beyond the Point Cloud: From Transductive to Semi-Supervised Learning Beyond the Point Cloud: From Transductive to Semi-Supervised Learning Vikas Sindhwani, Partha Niyogi, Mikhail Belkin Andrew B. Goldberg goldberg@cs.wisc.edu Department of Computer Sciences University of

More information

Fastfood Approximating Kernel Expansions in Loglinear Time. Quoc Le, Tamas Sarlos, and Alex Smola Presenter: Shuai Zheng (Kyle)

Fastfood Approximating Kernel Expansions in Loglinear Time. Quoc Le, Tamas Sarlos, and Alex Smola Presenter: Shuai Zheng (Kyle) Fastfood Approximating Kernel Expansions in Loglinear Time Quoc Le, Tamas Sarlos, and Alex Smola Presenter: Shuai Zheng (Kyle) Large Scale Problem: ImageNet Challenge Large scale data Number of training

More information

Gaussian Processes. Le Song. Machine Learning II: Advanced Topics CSE 8803ML, Spring 2012

Gaussian Processes. Le Song. Machine Learning II: Advanced Topics CSE 8803ML, Spring 2012 Gaussian Processes Le Song Machine Learning II: Advanced Topics CSE 8803ML, Spring 01 Pictorial view of embedding distribution Transform the entire distribution to expected features Feature space Feature

More information

Support Vector Machines (SVM) in bioinformatics. Day 1: Introduction to SVM

Support Vector Machines (SVM) in bioinformatics. Day 1: Introduction to SVM 1 Support Vector Machines (SVM) in bioinformatics Day 1: Introduction to SVM Jean-Philippe Vert Bioinformatics Center, Kyoto University, Japan Jean-Philippe.Vert@mines.org Human Genome Center, University

More information

Back to the future: Radial Basis Function networks revisited

Back to the future: Radial Basis Function networks revisited Back to the future: Radial Basis Function networks revisited Qichao Que, Mikhail Belkin Department of Computer Science and Engineering Ohio State University Columbus, OH 4310 que, mbelkin@cse.ohio-state.edu

More information

Multi-View Point Cloud Kernels for Semi-Supervised Learning

Multi-View Point Cloud Kernels for Semi-Supervised Learning Multi-View Point Cloud Kernels for Semi-Supervised Learning David S. Rosenberg, Vikas Sindhwani, Peter L. Bartlett, Partha Niyogi May 29, 2009 Scope In semi-supervised learning (SSL), we learn a predictive

More information

Stat542 (F11) Statistical Learning. First consider the scenario where the two classes of points are separable.

Stat542 (F11) Statistical Learning. First consider the scenario where the two classes of points are separable. Linear SVM (separable case) First consider the scenario where the two classes of points are separable. It s desirable to have the width (called margin) between the two dashed lines to be large, i.e., have

More information

Outline. Motivation. Mapping the input space to the feature space Calculating the dot product in the feature space

Outline. Motivation. Mapping the input space to the feature space Calculating the dot product in the feature space to The The A s s in to Fabio A. González Ph.D. Depto. de Ing. de Sistemas e Industrial Universidad Nacional de Colombia, Bogotá April 2, 2009 to The The A s s in 1 Motivation Outline 2 The Mapping the

More information

> DEPARTMENT OF MATHEMATICS AND COMPUTER SCIENCE GRAVIS 2016 BASEL. Logistic Regression. Pattern Recognition 2016 Sandro Schönborn University of Basel

> DEPARTMENT OF MATHEMATICS AND COMPUTER SCIENCE GRAVIS 2016 BASEL. Logistic Regression. Pattern Recognition 2016 Sandro Schönborn University of Basel Logistic Regression Pattern Recognition 2016 Sandro Schönborn University of Basel Two Worlds: Probabilistic & Algorithmic We have seen two conceptual approaches to classification: data class density estimation

More information

Each new feature uses a pair of the original features. Problem: Mapping usually leads to the number of features blow up!

Each new feature uses a pair of the original features. Problem: Mapping usually leads to the number of features blow up! Feature Mapping Consider the following mapping φ for an example x = {x 1,...,x D } φ : x {x1,x 2 2,...,x 2 D,,x 2 1 x 2,x 1 x 2,...,x 1 x D,...,x D 1 x D } It s an example of a quadratic mapping Each new

More information

Foundations of Deep Learning from a Kernel Point of View. Julien Mairal

Foundations of Deep Learning from a Kernel Point of View. Julien Mairal Foundations of Deep Learning from a Kernel Point of View Julien Mairal Inria Grenoble Berlin, CoSIP winter school, 2017 Julien Mairal Foundations of DL from a kernel point of view 1/124 1 Several Paradigms

More information

On non-parametric robust quantile regression by support vector machines

On non-parametric robust quantile regression by support vector machines On non-parametric robust quantile regression by support vector machines Andreas Christmann joint work with: Ingo Steinwart (Los Alamos National Lab) Arnout Van Messem (Vrije Universiteit Brussel) ERCIM

More information

Reproducing Kernel Hilbert Spaces

Reproducing Kernel Hilbert Spaces Reproducing Kernel Hilbert Spaces Lorenzo Rosasco 9.520 Class 03 February 9, 2011 About this class Goal In this class we continue our journey in the world of RKHS. We discuss the Mercer theorem which gives

More information

Learning parameters in ODEs

Learning parameters in ODEs Learning parameters in ODEs Application to biological networks Florence d Alché-Buc Joint work with Minh Quach and Nicolas Brunel IBISC FRE 3190 CNRS, Université d Évry-Val d Essonne, France /14 Florence

More information

Advances in kernel exponential families

Advances in kernel exponential families Advances in kernel exponential families Arthur Gretton Gatsby Computational Neuroscience Unit, University College London NIPS, 2017 1/39 Outline Motivating application: Fast estimation of complex multivariate

More information

Lecture 7: Kernels for Classification and Regression

Lecture 7: Kernels for Classification and Regression Lecture 7: Kernels for Classification and Regression CS 194-10, Fall 2011 Laurent El Ghaoui EECS Department UC Berkeley September 15, 2011 Outline Outline A linear regression problem Linear auto-regressive

More information

Statistical Learning Reading Assignments

Statistical Learning Reading Assignments Statistical Learning Reading Assignments S. Gong et al. Dynamic Vision: From Images to Face Recognition, Imperial College Press, 2001 (Chapt. 3, hard copy). T. Evgeniou, M. Pontil, and T. Poggio, "Statistical

More information

Lecture 10: Support Vector Machine and Large Margin Classifier

Lecture 10: Support Vector Machine and Large Margin Classifier Lecture 10: Support Vector Machine and Large Margin Classifier Applied Multivariate Analysis Math 570, Fall 2014 Xingye Qiao Department of Mathematical Sciences Binghamton University E-mail: qiao@math.binghamton.edu

More information

Lecture 18: Multiclass Support Vector Machines

Lecture 18: Multiclass Support Vector Machines Fall, 2017 Outlines Overview of Multiclass Learning Traditional Methods for Multiclass Problems One-vs-rest approaches Pairwise approaches Recent development for Multiclass Problems Simultaneous Classification

More information

Multi-view Metric Learning in Vector-valued Kernel Spaces

Multi-view Metric Learning in Vector-valued Kernel Spaces Riikka Huusari Hachem Kadri Cécile Capponi Aix Marseille Univ, Université de Toulon, CNRS, LIS, Marseille, France Abstract We consider the problem of metric learning for multi-view data and present a novel

More information

Approximation Theoretical Questions for SVMs

Approximation Theoretical Questions for SVMs Ingo Steinwart LA-UR 07-7056 October 20, 2007 Statistical Learning Theory: an Overview Support Vector Machines Informal Description of the Learning Goal X space of input samples Y space of labels, usually

More information

MIT 9.520/6.860, Fall 2017 Statistical Learning Theory and Applications. Class 19: Data Representation by Design

MIT 9.520/6.860, Fall 2017 Statistical Learning Theory and Applications. Class 19: Data Representation by Design MIT 9.520/6.860, Fall 2017 Statistical Learning Theory and Applications Class 19: Data Representation by Design What is data representation? Let X be a data-space X M (M) F (M) X A data representation

More information

Kernel Methods and Support Vector Machines

Kernel Methods and Support Vector Machines Kernel Methods and Support Vector Machines Oliver Schulte - CMPT 726 Bishop PRML Ch. 6 Support Vector Machines Defining Characteristics Like logistic regression, good for continuous input features, discrete

More information

Reproducing Kernel Hilbert Spaces

Reproducing Kernel Hilbert Spaces 9.520: Statistical Learning Theory and Applications February 10th, 2010 Reproducing Kernel Hilbert Spaces Lecturer: Lorenzo Rosasco Scribe: Greg Durrett 1 Introduction In the previous two lectures, we

More information

Less is More: Computational Regularization by Subsampling

Less is More: Computational Regularization by Subsampling Less is More: Computational Regularization by Subsampling Lorenzo Rosasco University of Genova - Istituto Italiano di Tecnologia Massachusetts Institute of Technology lcsl.mit.edu joint work with Alessandro

More information

Review: Support vector machines. Machine learning techniques and image analysis

Review: Support vector machines. Machine learning techniques and image analysis Review: Support vector machines Review: Support vector machines Margin optimization min (w,w 0 ) 1 2 w 2 subject to y i (w 0 + w T x i ) 1 0, i = 1,..., n. Review: Support vector machines Margin optimization

More information

Learning Good Representations for Multiple Related Tasks

Learning Good Representations for Multiple Related Tasks Learning Good Representations for Multiple Related Tasks Massimiliano Pontil Computer Science Dept UCL and ENSAE-CREST March 25, 2015 1 / 35 Plan Problem formulation and examples Analysis of multitask

More information

Support Vector Machine

Support Vector Machine Support Vector Machine Fabrice Rossi SAMM Université Paris 1 Panthéon Sorbonne 2018 Outline Linear Support Vector Machine Kernelized SVM Kernels 2 From ERM to RLM Empirical Risk Minimization in the binary

More information

Chap 1. Overview of Statistical Learning (HTF, , 2.9) Yongdai Kim Seoul National University

Chap 1. Overview of Statistical Learning (HTF, , 2.9) Yongdai Kim Seoul National University Chap 1. Overview of Statistical Learning (HTF, 2.1-2.6, 2.9) Yongdai Kim Seoul National University 0. Learning vs Statistical learning Learning procedure Construct a claim by observing data or using logics

More information

Learning to Learn and Collaborative Filtering

Learning to Learn and Collaborative Filtering Appearing in NIPS 2005 workshop Inductive Transfer: Canada, December, 2005. 10 Years Later, Whistler, Learning to Learn and Collaborative Filtering Kai Yu, Volker Tresp Siemens AG, 81739 Munich, Germany

More information

Kernel Methods. Jean-Philippe Vert Last update: Jan Jean-Philippe Vert (Mines ParisTech) 1 / 444

Kernel Methods. Jean-Philippe Vert Last update: Jan Jean-Philippe Vert (Mines ParisTech) 1 / 444 Kernel Methods Jean-Philippe Vert Jean-Philippe.Vert@mines.org Last update: Jan 2015 Jean-Philippe Vert (Mines ParisTech) 1 / 444 What we know how to solve Jean-Philippe Vert (Mines ParisTech) 2 / 444

More information

Metric Embedding for Kernel Classification Rules

Metric Embedding for Kernel Classification Rules Metric Embedding for Kernel Classification Rules Bharath K. Sriperumbudur University of California, San Diego (Joint work with Omer Lang & Gert Lanckriet) Bharath K. Sriperumbudur (UCSD) Metric Embedding

More information

Chapter 9. Support Vector Machine. Yongdai Kim Seoul National University

Chapter 9. Support Vector Machine. Yongdai Kim Seoul National University Chapter 9. Support Vector Machine Yongdai Kim Seoul National University 1. Introduction Support Vector Machine (SVM) is a classification method developed by Vapnik (1996). It is thought that SVM improved

More information

Bayesian Support Vector Machines for Feature Ranking and Selection

Bayesian Support Vector Machines for Feature Ranking and Selection Bayesian Support Vector Machines for Feature Ranking and Selection written by Chu, Keerthi, Ong, Ghahramani Patrick Pletscher pat@student.ethz.ch ETH Zurich, Switzerland 12th January 2006 Overview 1 Introduction

More information

Regularization in Reproducing Kernel Banach Spaces

Regularization in Reproducing Kernel Banach Spaces .... Regularization in Reproducing Kernel Banach Spaces Guohui Song School of Mathematical and Statistical Sciences Arizona State University Comp Math Seminar, September 16, 2010 Joint work with Dr. Fred

More information

Kernel Machines. Pradeep Ravikumar Co-instructor: Manuela Veloso. Machine Learning

Kernel Machines. Pradeep Ravikumar Co-instructor: Manuela Veloso. Machine Learning Kernel Machines Pradeep Ravikumar Co-instructor: Manuela Veloso Machine Learning 10-701 SVM linearly separable case n training points (x 1,, x n ) d features x j is a d-dimensional vector Primal problem:

More information

MIT 9.520/6.860, Fall 2018 Statistical Learning Theory and Applications. Class 04: Features and Kernels. Lorenzo Rosasco

MIT 9.520/6.860, Fall 2018 Statistical Learning Theory and Applications. Class 04: Features and Kernels. Lorenzo Rosasco MIT 9.520/6.860, Fall 2018 Statistical Learning Theory and Applications Class 04: Features and Kernels Lorenzo Rosasco Linear functions Let H lin be the space of linear functions f(x) = w x. f w is one

More information

Stability of Multi-Task Kernel Regression Algorithms

Stability of Multi-Task Kernel Regression Algorithms JMLR: Workshop and Conference Proceedings 29:1 16, 2013 ACML 2013 Stability of Multi-Task Kernel Regression Algorithms Julien Audiffren Hachem Kadri Aix-Marseille Université, CNRS, LIF UMR 7279, 13000,

More information

Lecture 3: Statistical Decision Theory (Part II)

Lecture 3: Statistical Decision Theory (Part II) Lecture 3: Statistical Decision Theory (Part II) Hao Helen Zhang Hao Helen Zhang Lecture 3: Statistical Decision Theory (Part II) 1 / 27 Outline of This Note Part I: Statistics Decision Theory (Classical

More information

An Introduction to Kernel Methods 1

An Introduction to Kernel Methods 1 An Introduction to Kernel Methods 1 Yuri Kalnishkan Technical Report CLRC TR 09 01 May 2009 Department of Computer Science Egham, Surrey TW20 0EX, England 1 This paper has been written for wiki project

More information

ECS289: Scalable Machine Learning

ECS289: Scalable Machine Learning ECS289: Scalable Machine Learning Cho-Jui Hsieh UC Davis Oct 27, 2015 Outline One versus all/one versus one Ranking loss for multiclass/multilabel classification Scaling to millions of labels Multiclass

More information

STATISTICAL BEHAVIOR AND CONSISTENCY OF CLASSIFICATION METHODS BASED ON CONVEX RISK MINIMIZATION

STATISTICAL BEHAVIOR AND CONSISTENCY OF CLASSIFICATION METHODS BASED ON CONVEX RISK MINIMIZATION STATISTICAL BEHAVIOR AND CONSISTENCY OF CLASSIFICATION METHODS BASED ON CONVEX RISK MINIMIZATION Tong Zhang The Annals of Statistics, 2004 Outline Motivation Approximation error under convex risk minimization

More information

Towards Deep Kernel Machines

Towards Deep Kernel Machines Towards Deep Kernel Machines Julien Mairal Inria, Grenoble Prague, April, 2017 Julien Mairal Towards deep kernel machines 1/51 Part I: Scientific Context Julien Mairal Towards deep kernel machines 2/51

More information

Support Vector Machines

Support Vector Machines Wien, June, 2010 Paul Hofmarcher, Stefan Theussl, WU Wien Hofmarcher/Theussl SVM 1/21 Linear Separable Separating Hyperplanes Non-Linear Separable Soft-Margin Hyperplanes Hofmarcher/Theussl SVM 2/21 (SVM)

More information

CLOSE-TO-CLEAN REGULARIZATION RELATES

CLOSE-TO-CLEAN REGULARIZATION RELATES Worshop trac - ICLR 016 CLOSE-TO-CLEAN REGULARIZATION RELATES VIRTUAL ADVERSARIAL TRAINING, LADDER NETWORKS AND OTHERS Mudassar Abbas, Jyri Kivinen, Tapani Raio Department of Computer Science, School of

More information

RegML 2018 Class 2 Tikhonov regularization and kernels

RegML 2018 Class 2 Tikhonov regularization and kernels RegML 2018 Class 2 Tikhonov regularization and kernels Lorenzo Rosasco UNIGE-MIT-IIT June 17, 2018 Learning problem Problem For H {f f : X Y }, solve min E(f), f H dρ(x, y)l(f(x), y) given S n = (x i,

More information

PREDICTING SOLAR GENERATION FROM WEATHER FORECASTS. Chenlin Wu Yuhan Lou

PREDICTING SOLAR GENERATION FROM WEATHER FORECASTS. Chenlin Wu Yuhan Lou PREDICTING SOLAR GENERATION FROM WEATHER FORECASTS Chenlin Wu Yuhan Lou Background Smart grid: increasing the contribution of renewable in grid energy Solar generation: intermittent and nondispatchable

More information

A Least Squares Formulation for Canonical Correlation Analysis

A Least Squares Formulation for Canonical Correlation Analysis A Least Squares Formulation for Canonical Correlation Analysis Liang Sun, Shuiwang Ji, and Jieping Ye Department of Computer Science and Engineering Arizona State University Motivation Canonical Correlation

More information

Max Margin-Classifier

Max Margin-Classifier Max Margin-Classifier Oliver Schulte - CMPT 726 Bishop PRML Ch. 7 Outline Maximum Margin Criterion Math Maximizing the Margin Non-Separable Data Kernels and Non-linear Mappings Where does the maximization

More information

Part 2: Multivariate fmri analysis using a sparsifying spatio-temporal prior

Part 2: Multivariate fmri analysis using a sparsifying spatio-temporal prior Chalmers Machine Learning Summer School Approximate message passing and biomedicine Part 2: Multivariate fmri analysis using a sparsifying spatio-temporal prior Tom Heskes joint work with Marcel van Gerven

More information

Hilbert Space Methods in Learning

Hilbert Space Methods in Learning Hilbert Space Methods in Learning guest lecturer: Risi Kondor 6772 Advanced Machine Learning and Perception (Jebara), Columbia University, October 15, 2003. 1 1. A general formulation of the learning problem

More information

Lecture 5: GPs and Streaming regression

Lecture 5: GPs and Streaming regression Lecture 5: GPs and Streaming regression Gaussian Processes Information gain Confidence intervals COMP-652 and ECSE-608, Lecture 5 - September 19, 2017 1 Recall: Non-parametric regression Input space X

More information

Model Selection for Gaussian Processes

Model Selection for Gaussian Processes Institute for Adaptive and Neural Computation School of Informatics,, UK December 26 Outline GP basics Model selection: covariance functions and parameterizations Criteria for model selection Marginal

More information

Kernel Logistic Regression and the Import Vector Machine

Kernel Logistic Regression and the Import Vector Machine Kernel Logistic Regression and the Import Vector Machine Ji Zhu and Trevor Hastie Journal of Computational and Graphical Statistics, 2005 Presented by Mingtao Ding Duke University December 8, 2011 Mingtao

More information

Introduction to Machine Learning Midterm, Tues April 8

Introduction to Machine Learning Midterm, Tues April 8 Introduction to Machine Learning 10-701 Midterm, Tues April 8 [1 point] Name: Andrew ID: Instructions: You are allowed a (two-sided) sheet of notes. Exam ends at 2:45pm Take a deep breath and don t spend

More information

Bayesian Aggregation for Extraordinarily Large Dataset

Bayesian Aggregation for Extraordinarily Large Dataset Bayesian Aggregation for Extraordinarily Large Dataset Guang Cheng 1 Department of Statistics Purdue University www.science.purdue.edu/bigdata Department Seminar Statistics@LSE May 19, 2017 1 A Joint Work

More information

Statistical Machine Learning from Data

Statistical Machine Learning from Data Samy Bengio Statistical Machine Learning from Data 1 Statistical Machine Learning from Data Support Vector Machines Samy Bengio IDIAP Research Institute, Martigny, Switzerland, and Ecole Polytechnique

More information