Max-Margin Latent Variable Models M. Pawan Kumar.

Slides:



Advertisements
Similar presentations
Self-Paced Learning for Semantic Segmentation
Advertisements

Poselets: Body Part Detectors trained Using 3D Human Pose Annotations Lubomir Bourdev & Jitendra Malik ICCV 2009.
Solving Markov Random Fields using Second Order Cone Programming Relaxations M. Pawan Kumar Philip Torr Andrew Zisserman.
Latent Variables Naman Agarwal Michael Nute May 1, 2013.
Weakly supervised learning of MRF models for image region labeling Jakob Verbeek LEAR team, INRIA Rhône-Alpes.
Learning Specific-Class Segmentation from Diverse Data M. Pawan Kumar, Haitherm Turki, Dan Preston and Daphne Koller at ICCV 2011 VGG reading group, 29.
Learning Shared Body Plans Ian Endres University of Illinois work with Derek Hoiem, Vivek Srikumar and Ming-Wei Chang.
Curriculum Learning for Latent Structural SVM
Unsupervised Learning Clustering K-Means. Recall: Key Components of Intelligent Agents Representation Language: Graph, Bayes Nets, Linear functions Inference.
Ľubor Ladický1 Phil Torr2 Andrew Zisserman1
Efficient Large-Scale Structured Learning
Structured SVM Chen-Tse Tsai and Siddharth Gupta.
Loss-based Visual Learning with Weak Supervision M. Pawan Kumar Joint work with Pierre-Yves Baudin, Danny Goodman, Puneet Kumar, Nikos Paragios, Noura.
Ivan Laptev IRISA/INRIA, Rennes, France September 07, 2006 Boosted Histograms for Improved Object Detection.
Lecture 31: Modern object recognition
Many slides based on P. FelzenszwalbP. Felzenszwalb General object detection with deformable part-based models.
Structural Human Action Recognition from Still Images Moin Nabi Computer Vision Lab. ©IPM - Oct
Learning Structural SVMs with Latent Variables Xionghao Liu.
Object-centric spatial pooling for image classification Olga Russakovsky, Yuanqing Lin, Kai Yu, Li Fei-Fei ECCV 2012.
Enhancing Exemplar SVMs using Part Level Transfer Regularization 1.
More sliding window detection: Discriminative part-based models Many slides based on P. FelzenszwalbP. Felzenszwalb.
Jun Zhu Dept. of Comp. Sci. & Tech., Tsinghua University This work was done when I was a visiting researcher at CMU. Joint.
Restrict learning to a model-dependent “easy” set of samples General form of objective: Introduce indicator of “easiness” v i : K determines threshold.
Learning to Segment with Diverse Data M. Pawan Kumar Stanford University.
Poselets Michael Krainin CSE 590V Oct 18, Person Detection Dalal and Triggs ‘05 – Learn to classify pedestrians vs. background – HOG + linear SVM.
Generic Object Detection using Feature Maps Oscar Danielsson Stefan Carlsson
Learning Spatial Context: Using stuff to find things Geremy Heitz Daphne Koller Stanford University October 13, 2008 ECCV 2008.
Learning to Segment from Diverse Data M. Pawan Kumar Daphne KollerHaithem TurkiDan Preston.
Learning Spatial Context: Can stuff help us find things? Geremy Heitz Daphne Koller April 14, 2008 DAGS Stuff (n): Material defined by a homogeneous or.
What, Where & How Many? Combining Object Detectors and CRFs
Lecture 29: Recent work in recognition CS4670: Computer Vision Noah Snavely.
Generic object detection with deformable part-based models
Ranking with High-Order and Missing Information M. Pawan Kumar Ecole Centrale Paris Aseem BehlPuneet DokaniaPritish MohapatraC. V. Jawahar.
Modeling Latent Variable Uncertainty for Loss-based Learning Daphne Koller Stanford University Ben Packer Stanford University M. Pawan Kumar École Centrale.
Loss-based Learning with Weak Supervision M. Pawan Kumar.
Self-paced Learning for Latent Variable Models
Loss-based Learning with Latent Variables M. Pawan Kumar École Centrale Paris École des Ponts ParisTech INRIA Saclay, Île-de-France Joint work with Ben.
Computer Vision CS 776 Spring 2014 Recognition Machine Learning Prof. Alex Berg.
Ranking with High-Order and Missing Information M. Pawan Kumar Ecole Centrale Paris Aseem BehlPuneet KumarPritish MohapatraC. V. Jawahar.
Learning a Small Mixture of Trees M. Pawan Kumar Daphne Koller Aim: To efficiently learn a.
Modeling Latent Variable Uncertainty for Loss-based Learning Daphne Koller Stanford University Ben Packer Stanford University M. Pawan Kumar École Centrale.
Optimizing Average Precision using Weakly Supervised Data Aseem Behl IIIT Hyderabad Under supervision of: Dr. M. Pawan Kumar (INRIA Paris), Prof. C.V.
Lecture 31: Modern recognition CS4670 / 5670: Computer Vision Noah Snavely.
Deformable Part Models (DPM) Felzenswalb, Girshick, McAllester & Ramanan (2010) Slides drawn from a tutorial By R. Girshick AP 12% 27% 36% 45% 49% 2005.
Tell Me What You See and I will Show You Where It Is Jia Xu 1 Alexander G. Schwing 2 Raquel Urtasun 2,3 1 University of Wisconsin-Madison 2 University.
Object detection, deep learning, and R-CNNs
Multiple Instance Learning for Sparse Positive Bags Razvan C. Bunescu Machine Learning Group Department of Computer Sciences University of Texas at Austin.
Pictorial Structures and Distance Transforms Computer Vision CS 543 / ECE 549 University of Illinois Ian Endres 03/31/11.
Improved Object Detection
Recognition Using Visual Phrases
Learning from Big Data Lecture 5
Maximum Entropy Discrimination Tommi Jaakkola Marina Meila Tony Jebara MIT CMU MIT.
Object Recognition by Integrating Multiple Image Segmentations Caroline Pantofaru, Cordelia Schmid, Martial Hebert ECCV 2008 E.
Describing People: A Poselet-Based Approach to Attribute Classification.
Optimizing Average Precision using Weakly Supervised Data Aseem Behl 1, C.V. Jawahar 1 and M. Pawan Kumar 2 1 IIIT Hyderabad, India, 2 Ecole Centrale Paris.
Loss-based Learning with Weak Supervision M. Pawan Kumar.
Sreekanth Vempati ( ) Advisors: Dr. C. V. Jawahar ( IIIT Hyderabad ), Dr. Andrew Zisserman ( Univ. of Oxford ) Efficient SVM based object classification.
Parsing Natural Scenes and Natural Language with Recursive Neural Networks INTERNATIONAL CONFERENCE ON MACHINE LEARNING (ICML 2011) RICHARD SOCHER CLIFF.
1 Bilinear Classifiers for Visual Recognition Computational Vision Lab. University of California Irvine To be presented in NIPS 2009 Hamed Pirsiavash Deva.
Discriminative Machine Learning Topic 4: Weak Supervision M. Pawan Kumar Slides available online
CS 4501: Introduction to Computer Vision Object Localization, Detection, Semantic Segmentation Connelly Barnes Some slides from Fei-Fei Li / Andrej Karpathy.
Learning a Region-based Scene Segmentation Model
Object detection with deformable part-based models
Object Localization Goal: detect the location of an object within an image Fully supervised: Training data labeled with object category and ground truth.
Efficiently Selecting Regions for Scene Understanding
Group Norm for Learning Latent Structural SVMs
Self-Paced Learning for Semisupervised Image Classification
Multiple Feature Learning for Action Classification
Cascaded Classification Models
Presentation transcript:

Max-Margin Latent Variable Models M. Pawan Kumar

Max-Margin Latent Variable Models M. Pawan Kumar Daphne Koller Ben Packer Kevin Miller, Rafi Witten, Tim Tang, Danny Goodman, Haithem Turki, Dan Preston, Dan Selsam, Andrej Karpathy

Computer Vision Data Segmentation Information Log (Size) ~ 2000

Computer Vision Data Segmentation Log (Size) Bounding Box ~ 2000 ~ Information

Computer Vision Data Segmentation Log (Size) Bounding Box Image-Level ~ 2000 ~ > 14 M “Car” “Chair” Information

Computer Vision Data Segmentation Log (Size) Bounding Box Image-Level Noisy Label ~ 2000 ~ > 14 M > 6 B Learn with missing information (latent variables) Information

Two Types of Problems Latent SVM (Background) Self-Paced Learning Max-Margin Min-Entropy Models Discussion Outline

Annotation Mismatch Learn to classify an image Image x Annotation a = “Deer” Mismatch between desired and available annotations h Exact value of latent variable is not “important”

Annotation Mismatch Learn to classify a DNA sequence Mismatch between desired and possible annotations Exact value of latent variable is not “important” Sequence x Annotation a  {+1, -1} Latent Variables h

Output Mismatch Learn to segment an image Image xOutput y

Output Mismatch Learn to segment an image Bird (x, a) (a, h)

Output Mismatch Learn to segment an image Mismatch between desired output and available annotations Exact value of latent variable is important (x, a) (a, h) Cow

Output Mismatch Learn to classify actions (x, y)

Output Mismatch Learn to classify actions + “jumping” xh a = +1 hbhb

Output Mismatch Learn to classify actions + “jumping” xh a = -1 hbhb Mismatch between desired output and available annotations Exact value of latent variable is important

Two Types of Problems Latent SVM (Background) Self-Paced Learning Max-Margin Min-Entropy Models Discussion Outline

Latent SVM Features  (x,a,h) wT(x,a,h)wT(x,a,h) Parameters w Image x Annotation a = “Deer” h Andrews et al, 2001; Smola et al, 2005; Felzenszwalb et al, 2008; Yu and Joachims, 2009 (a(w),h(w)) = max a,h

Parameter Learning Score of Ground-Truth > Score of All Other Outputs Best Completion of

Parameter Learning max h w T  (x i,a i,h) > wT(x,a,h)wT(x,a,h)

Parameter Learning max h w T  (x i,a i,h) ≥ wT(x,a,h)wT(x,a,h) + Δ(a i,a) - ξ i min ||w|| 2 + CΣ i ξ i Annotation Mismatch

Optimization Update h i * = argmax h w T  (x i,a i,h) Update w by solving a convex problem min ||w|| 2 + C∑ i  i w T  (x i,a i,h i *) - w T  (x i,a,h) ≥  (a i, a) -  i Repeat until convergence

Two Types of Problems Latent SVM (Background) Self-Paced Learning Max-Margin Min-Entropy Models Discussion Outline

Self-Paced Learning Kumar, Packer and Koller, NIPS = 2 1/3 + 1/6 = 1/2 e iπ +1 = 0 Math is for losers !! FAILURE … BAD LOCAL MINIMUM

Self-Paced Learning Kumar, Packer and Koller, NIPS 2010 Euler was a Genius!! SUCCESS … GOOD LOCAL MINIMUM = 2 1/3 + 1/6 = 1/2 e iπ +1 = 0

Optimization Update h i * = argmax h w T  (x i,a i,h) Update w by solving a convex problem min ||w|| 2 + C∑ i  i Repeat until convergence vivi v i  {0,1} λ  λμλ  λμ - λ∑ i v i w T  (x i,a i,h i *) - w T  (x i,a,h) ≥  (a i, a) -  i

Image Classification 271 images, 6 classes 90/10 train/test split 5 folds Mammals Dataset

Image Classification Kumar, Packer and Koller, NIPS 2010 CCCP SPL CCCP SPL HOG-Based Model. Dalal and Triggs, 2005

Image Classification ~ 5000 images 50/50 train/test split 5 folds PASCAL VOC 2007 Dataset Car vs. Not-Car

Image Classification Witten, Miller, Kumar, Packer and Koller, In Preparation Objective HOG + Dense SIFT + Dense Color SIFT SPL+ – Different features choose different “easy” samples

Image Classification Witten, Miller, Kumar, Packer and Koller, In Preparation Mean Average Precision HOG + Dense SIFT + Dense Color SIFT SPL+ – Different features choose different “easy” samples

Motif Finding ~ 40,000 sequences 50/50 train/test split 5 folds UniProbe Dataset Binding vs. Not-Binding

Motif Finding Kumar, Packer and Koller, NIPS 2010 CCCP SPL CCCP SPL Motif + Markov Background Model. Yu and Joachims, 2009

Semantic Segmentation + Train images Validation - 53 images Test - 90 images Train images Validation images Test images Stanford BackgroundVOC Segmentation 2009

Semantic Segmentation ImageNetVOC Detection Train imagesTrain images Bounding Box Data Image-Level Data

Semantic Segmentation Kumar, Turki, Preston and Koller, ICCV 2011 SUP CCCP SPL SUP CCCP SPL Region-based Model. Gould, Fulton and Koller, 2009 SUP – Supervised Learning (Segmentation Data Only)

Action Classification PASCAL VOC 2011 Train – 3000 instances Train images Bounding Box Data Noisy Data + Test – 3000 instances

Action Classification Packer, Kumar, Tang and Koller, In Preparation SUP CCCP SPL Poselet-based Model. Maji, Bourdev and Malik, 2011

Self-Paced Multiple Kernel Learning Kumar, Packer and Koller, In Preparation = 2 1/3 + 1/6 = 1/2 e iπ +1 = 0 Integers Rational Numbers Imaginary Numbers USE A FIXED MODEL

Kumar, Packer and Koller, In Preparation = 2 1/3 + 1/6 = 1/2 e iπ +1 = 0 Integers Rational Numbers Imaginary Numbers ADAPT THE MODEL COMPLEXITY Self-Paced Multiple Kernel Learning

Optimization Update h i * = argmax h w T  (x i,a i,h) Update w by solving a convex problem min ||w|| 2 + C∑ i  i Repeat until convergence vivi v i  {0,1} λ  λμλ  λμ - λ∑ i v i w T  (x i,a i,h i *) - w T  (x i,a,h) ≥  (a i, a) -  i K ij =  (x i,a i,h i ) T  (x j,a j,h j ) K = Σ k c k K k ^ and c

Image Classification 271 images, 6 classes 90/10 train/test split 5 folds Mammals Dataset

Image Classification Kumar, Packer and Koller, In Preparation FIXED SPMKL FIXED SPMKL HOG-Based Model. Dalal and Triggs, 2005

Motif Finding ~ 40,000 sequences 50/50 train/test split 5 folds UniProbe Dataset Binding vs. Not-Binding

Motif Finding Kumar, Packer and Koller, NIPS 2010 FIXED SPMKL FIXED SPMKL Motif + Markov Background Model. Yu and Joachims, 2009

Two Types of Problems Latent SVM (Background) Self-Paced Learning Max-Margin Min-Entropy Models Discussion Outline

Pr(a,h|x) = exp( w T  (x,a,h)) Z(x) Pr(a 1,h|x) MAP Inference

Pr(a 1,h|x) Pr(a 2,h|x) MAP Inference min a,h – log (Pr(a,h|x)) Value of latent variable? Pr(a,h|x) = exp( w T  (x,a,h)) Z(x)

min a – log (Pr(a|x)) Min-Entropy Inference + H α (Pr(h|a,x)) min a H α (Q(a; x, w)) Q(a; x, w) = Set of all {Pr(a,h|x)} Renyi entropy of generalized distribution

min ||w|| 2 + C∑ i  i H α (Q(a; x, w))- H α (Q(a i ; x, w)) ≥  (a i, a) -  i  i ≥ 0 Like latent SVM, minimizes  (a i, a i (w)) In fact, when α = ∞... Max-Margin Min-Entropy Models Miller, Kumar, Packer, Goodman and Koller, AISTATS 2012

min ||w|| 2 + C∑ i  i max h w T  (x,a i,h)-max h w T  (x,a,h) ≥  (a i, a) -  i  i ≥ 0 In fact, when α = ∞... Latent SVM Max-Margin Min-Entropy Models Like latent SVM, minimizes  (a i, a i (w)) Miller, Kumar, Packer, Goodman and Koller, AISTATS 2012

Image Classification 271 images, 6 classes 90/10 train/test split 5 folds Mammals Dataset

Image Classification Miller, Kumar, Packer, Goodman and Koller, AISTATS 2012 HOG-Based Model. Dalal and Triggs, 2005

Image Classification Miller, Kumar, Packer, Goodman and Koller, AISTATS 2012 HOG-Based Model. Dalal and Triggs, 2005

Image Classification Miller, Kumar, Packer, Goodman and Koller, AISTATS 2012 HOG-Based Model. Dalal and Triggs, 2005

Motif Finding ~ 40,000 sequences 50/50 train/test split 5 folds UniProbe Dataset Binding vs. Not-Binding

Motif Finding Miller, Kumar, Packer, Goodman and Koller, AISTATS 2012 Motif + Markov Background Model. Yu and Joachims, 2009

Two Types of Problems Latent SVM (Background) Self-Paced Learning Max-Margin Min-Entropy Models Discussion Outline

Very Large Datasets Initialize parameters using supervised data Impute latent variables (inference) Select easy samples (very efficient) Update parametersusing incremental SVM Refine efficiently with proximal regularization

Output Mismatch Δ(a,h,a(w),h(w)) Σ h Pr θ (h|a,x)+ A(θ) C. R. Rao’s Relative Quadratic Entropy Minimize over w and θ

Output Mismatch Δ(a,h,a(w),h(w)) Σ h Pr θ (h|a,x)+ A(θ) C. R. Rao’s Relative Quadratic Entropy Minimize over w (a 1,h) (a 2,h) Pr θ (h,a|x)

Output Mismatch Δ(a,h,a(w),h(w)) Σ h Pr θ (h|a,x)+ A(θ) C. R. Rao’s Relative Quadratic Entropy Minimize over w (a 1,h) Pr θ (h,a|x) (a 2,h)

Output Mismatch Δ(a,h,a(w),h(w)) Σ h Pr θ (h|a,x)+ A(θ) C. R. Rao’s Relative Quadratic Entropy Minimize over θ (a 1,h) (a 2,h) Pr θ (h,a|x)

Output Mismatch Δ(a,h,a(w),h(w)) Σ h Pr θ (h|a,x)+ A(θ) C. R. Rao’s Relative Quadratic Entropy Minimize over θ (a 1,h) (a 2,h) Pr θ (h,a|x)

Output Mismatch Δ(a,h,a(w),h(w)) Σ h Pr θ (h|a,x)+ A(θ) C. R. Rao’s Relative Quadratic Entropy Minimize over θ (a 1,h) (a 2,h) Pr θ (h,a|x)

Questions?