Alina Ene. From Minimum Cut to Submodular Minimization Leveraging the Decomposable Structure. Boston University
|
|
- Mariah Weaver
- 6 years ago
- Views:
Transcription
1 From Minimum Cut to Submodular Minimization Leveraging the Decomposable Structure Alina Ene Boston University Joint work with Huy Nguyen (Northeastern University) Laszlo Vegh (London School of Economics)
2 Minimum s-t Cut Image Segmentation s t x x 1 Submodular Minimization min f(s) S V x x 4 x 3 x 5 Structured Sparsity [Bach 10] MAP Inference
3 Minimum s-t Cut Image Segmentation s Submodular Minimization t min f(s) S V x x 1 Submodular Minimization min f(s) S V x x 4 x 3 x 5 Structured Sparsity [Bach 10] MAP Inference
4 Minimum s-t Cut Image Segmentation s Submodular Minimization t min f(s) S V Solvable in polynomial time x Combinatorial, Submodular ellipsoid, Minimization cutting plane, x 1 min f(s) S V x x 4 x 3 x 5 Structured Sparsity [Bach 10] MAP Inference
5 Minimum s-t Cut Image Segmentation s Submodular Minimization t min f(s) S V Solvable in polynomial time x Combinatorial, Submodular ellipsoid, Minimization cutting plane, x 1 min f(s) S V Very high running times ( V 3 ) x x 4 Leverage structure to obtain faster algos x 3 x 5 Structured Sparsity [Bach 10] MAP Inference
6 Decomposable Functions min S V f i (S) Simple fi : fast subroutine for minimizing fi(s) + w(s) for any linear function w [Stobbe, Krause 10; Kolmogorov 1; Jegelka, Bach, Sra 13; Nishihara, Jegelka, Jordan 14]
7 Decomposable Structure = + + f(s) = X ee f e (S) Cut function of a single edge
8 Decomposable Structure = + + hyperedges, local functions,
9 This Talk Sum-of-Submodular / Decomposable SFM min S V f i (S) Continuous algos based on gradient descent Combinatorial algos based on max flow approaches Interplay between continuous optimization and combinatorial structure
10 Main Combinatorial Player Key concept underpinning submodular algorithms Base Polytope B(f) ={y hy, 1 S iapplef(s) 8S V, hy, 1 V i = f(v )} Think of y as a linear function on V
11 Continuous Algorithms
12 From Submodular to Convex Optimization [Edmonds 70, Lovasz 83, Fujishige 84] Submodular Minimization can be reduced to finding a point of minimum norm in the base polytope Discrete Problem min f(s) S V Exact Convex Formulation min yb(f) kyk
13 Discrete Problem min f(s) S V Exact Convex Formulation min yb(f) kyk Convex objective is very nice Constraint is very complicated
14 Discrete Problem min f(s) S V Exact Convex Formulation min yb(f) kyk Convex objective is very nice Constraint is very complicated Algorithms based on Ellipsoid/Cutting Plane [GLS 84, LSW 15] Provable polynomial running times Running time prohibitively large in applications
15 Discrete Problem min f(s) S V Exact Convex Formulation min yb(f) kyk Convex objective is very nice Constraint is very complicated Fujishige-Wolfe and Frank-Wolfe algorithms Suitable for some applications Pseudo-polynomial running times Known to require large number of iterations
16 In the decomposable setting, can leverage more of the convex optimization toolkit Lemma: If f = f i then B(f) = B(f i ) Discrete Problem f i (S) min S V Exact Convex Formulation min B(f i ) f i simple ) quadratic min over B(f i ) easy
17 Discrete Problem f i (S) min S V Exact Convex Formulation min B(f i ) Decomposition allows for fast and practical algorithms based on gradient descent [Stobbe, Krause 10; Jegelka, Bach, Sra 13; Nishihara, Jegelka, Jordan 14, E., Nguyen 15]
18 Discrete Problem f i (S) min S V Exact Convex Formulation min B(f i ) [E., Nguyen ICML 15] Minimize the convex formulation using Random Coordinate Gradient Descent Also accelerated algorithm Fastest running times currently known
19 Exact Convex Formulation min B(f i ) g(y) := Random Coordinate Descent Algorithm Start with y =(y 1,...,y r ), where B(f i ) In each iteration Pick an index i {1,,...,r} uniformly at random hhupdate block iii y 0 i g(y)+hr argmin i g(y),z i i + kz i k z i B(f i )
20 Smoothness g kr i g(y) r i g(z)k applek z i k y and z differ onln block i Random Coordinate Descent Algorithm Start with y =(y 1,...,y r ), where B(f i ) In each iteration Pick an index i {1,,...,r} uniformly at random hhupdate block iii y 0 i g(y)+hr argmin i g(y),z i i + kz i k z i B(f i )
21 g Smooth functions have quadratic upper bounds y Random Coordinate Descent Algorithm Start with y =(y 1,...,y r ), where B(f i ) In each iteration Pick an index i {1,,...,r} uniformly at random hhupdate block iii y 0 i g(y)+hr argmin i g(y),z i i + kz i k z i B(f i )
22 (z) =g(y)+hr i g(y),z i i + kz i k z and y differ onln block i g y Random Coordinate Descent Algorithm Start with y =(y 1,...,y r ), where B(f i ) In each iteration Pick an index i {1,,...,r} uniformly at random hhupdate block iii y 0 i g(y)+hr argmin i g(y),z i i + kz i k z i B(f i )
23 (z) =g(y)+hr i g(y),z i i + kz i k z and y differ onln block i g Minimize the upper bound y y Random Coordinate Descent Algorithm Start with y =(y 1,...,y r ), where B(f i ) In each iteration Pick an index i {1,,...,r} uniformly at random hhupdate block iii y 0 i g(y)+hr argmin i g(y),z i i + kz i k z i B(f i )
24 Exact Convex Formulation min B(f i ) g(y) := Random Coordinate Descent Algorithm Start with y =(y 1,...,y r ), where B(f i ) In each iteration Pick an index i {1,,...,r} uniformly at random hhupdate block iii y 0 i g(y)+hr argmin i g(y),z i i + kz i k z i B(f i ) Efficient update for simple functions
25 Exact Convex Formulation min B(f i ) g(y) := Random Coordinate Descent Algorithm Start with y =(y 1,...,y r ), where B(f i ) In each iteration Pick an index i {1,,...,r} uniformly at random hhupdate block iii y 0 i g(y)+hr argmin i g(y),z i i + kz i k z i B(f i ) Also accelerated version building on [Fercoq-Richtárik 13]
26 Discrete Problem f i (S) min S V Exact Convex Formulation min B(f i ) [E., Nguyen ICML 15] Minimize the convex formulation using Random Coordinate Gradient Descent
27 Discrete Problem f i (S) min S V Exact Convex Formulation min B(f i ) [E., Nguyen ICML 15] Minimize the convex formulation using Random Coordinate Gradient Descent Running time (with acceleration) [ENV 17] O nr log Fmax Q Q: time for one update via oracle for fi
28 Discrete Problem f i (S) min S V Exact Convex Formulation min B(f i ) [E., Nguyen ICML 15] Minimize the convex formulation using Random Coordinate Gradient Descent Running time (with acceleration) [ENV 17] O(nr log (nf max ) Q) Q: time for one update via oracle for fi
29 Discrete Algorithms
30 The quest for combinatorial algos for submodular min [Cunningham 85] It is an outstanding open problem to find a practical combinatorial algorithm to minimize a general submodular function, which also runs in polynomial time.
31 The quest for combinatorial algos for submodular min [Cunningham 85] It is an outstanding open problem to find a practical combinatorial algorithm to minimize a general submodular function, which also runs in polynomial time. [Bixby et al. 85, Cunningham 85] Framework for designing combinatorial algorithms for submodular minimization
32 Basic framework for all combinatorial algorithms Cunningham Framework Maintain a point n the base polytope Iteratively update y using flow-style updates How to represent a point in the base polytope?
33 Basic framework for all combinatorial algorithms Cunningham Framework Maintain a point n the base polytope Iteratively update y using flow-style updates How to represent a point in the base polytope? As a convex combination of vertices Pro: Very general (works for any submodular fn) Con: Difficult/expensive to maintain and update
34 For decomposable submodular functions simpler version of Cunningham s approach works! Recall: For f = we have f i B(f) = B(f i ) Can represent y B(f) as y = y y r where B(f i ) Verify B(f i ) using oracle for f i
35 From decomposable functions to graphs Maintain a point y = y y r y B(f) into points with a decomposition B(f i ) Update using auxiliary graph G =(V,[ i E i ) (u, v) E i if y i B(f i ) for some > 0 c(u, v) = max u v u y i + v For graph cut instance, the auxiliary graph is the usual residual graph for network flows
36 Translate classical maxflow toolkit to submodular Edmonds-Karp-Dinitz, Preflow-push, SUBMODULAR
37 Translate classical maxflow toolkit to submodular Edmonds-Karp-Dinitz, Preflow-push, Sample Result: If all f i O(n r) have small support preflow-push runs in oracle calls First shown by [Fujishige-Zhang 94] for r = [Fix et al. 13] adapt IBFS algo. of [Goldberg et al. 11]
38 Translate classical maxflow toolkit to submodular Edmonds-Karp-Dinitz, Preflow-push, [E., Nguyen, Vegh NIPS 17] Discrete to Continuous Tight running time analyses for the gradient descent algorithms Discrete vs Continuous Experimental comparison on computer vision tasks
39 Recall our problem: min 8 < : g(y) = : y ry 9 = B(f i ) ; optimal solutions ry B(f i ) Minimize objective using random coordinate descent Linear convergence rate despite lack of strong convexity Use combinatorial structure instead of strong convexity
40 min n f(x): x X o µ-strong Convexity X f( ) f( ) hrf( ), i + µ (d(, ) ) for any, X
41 min n f(x): x X o µ-strong Convexity X f( ) f( ) hrf( ), i + µ (d(, ) ) for any, X What we really need f( ) f( ) µ (min d(, ) ) where is an optimal point
42 Our problem: min 8 < : g(y) = : y ry 9 = B(f i ) ; ry B(f i ) optimal point arbitrary point Restricted µ -Strong Convexity g( ) g( ) µ (min d(, ) )
43 Our problem: min 8 < : g(y) = : y ry 9 = B(f i ) ; ry B(f i ) optimal point arbitrary point Restricted µ -Strong Convexity g( ) g( ) µ (min d(, ) ) Can show µ max n 1 nr, 1 n o
44 Our problem: min 8 < : g(y) = : y ry 9 = B(f i ) ; ry B(f i ) Want: g( ) - g( ) 1 n d(, )
45 Our problem: min 8 < : g(y) = : y ry 9 = B(f i ) ; ry B(f i ) B(f) sum of the blocks Want: g( ) - g( ) 1 n d(, )
46 Our problem: min 8 < : g(y) = : y ry 9 = B(f i ) ; min norm point ry B(f i ) B(f) sum of the blocks Want: g( ) - g( ) 1 n d(, )
47 Our problem: min 8 < : g(y) = : y ry 9 = B(f i ) ; min norm point ry B(f i ) B(f) sum of the blocks g( ) - g( ) d(, ) 1 n d(, )
48 Our problem: min 8 < : g(y) = : y ry 9 = B(f i ) ; min norm point ry B(f i ) B(f) sum of the blocks g( ) - g( ) 1 d(, ) d(, ) n
49 Our problem: min 8 < : g(y) = : y ry 9 = B(f i ) ; flow decomposition min norm point ry B(f i ) B(f) sum of the blocks g( ) - g( ) 1 d(, ) d(, ) n
50 demand 1 n demand 1 n ` of demand = 1 n n ` of flow = n ` of demand = 1 n (` of flow)
51 Summary Graph Min Cut Decomposable Submod Min Algorithms that leverage both convex optimization and the combinatorial structure (via graph max flow)
Inference in Higher Order MRF-MAP Problems with Small and Large Cliques
Inference in Higher Order MRF-MAP Problems with Small and Large Cliques Ishant Shanu 1 Chetan Arora 1 S.N. Maheshwari 2 1 IIIT Delhi 2 IIT Delhi Abstract Higher Order MRF-MAP formulation has been a popular
More informationConditional gradient algorithms for machine learning
000 001 002 003 004 005 006 007 008 009 010 011 012 013 014 015 016 017 018 019 020 021 022 023 024 025 026 027 028 029 030 031 032 033 034 035 036 037 038 039 040 041 042 043 044 045 046 047 048 049 050
More informationStructured Optimal Transport
Structured Optimal Transport David Alvarez-Melis, Tommi Jaakkola, Stefanie Jegelka CSAIL, MIT OTML Workshop @ NIPS, Dec 9th 2017 Motivation: Domain Adaptation c(x i,y j ) c(x k,y`) Labeled Source Domain
More informationCombinatorial Selection and Least Absolute Shrinkage via The CLASH Operator
Combinatorial Selection and Least Absolute Shrinkage via The CLASH Operator Volkan Cevher Laboratory for Information and Inference Systems LIONS / EPFL http://lions.epfl.ch & Idiap Research Institute joint
More informationOptimization for Machine Learning
Optimization for Machine Learning (Problems; Algorithms - C) SUVRIT SRA Massachusetts Institute of Technology PKU Summer School on Data Science (July 2017) Course materials http://suvrit.de/teaching.html
More informationThe Min Norm Point on a Convex Polytope
The Min Norm Point on a Convex Polytope Jamie Haddock January 13, 2018 Graduate Group in Applied Mathematics UC Davis joint with Jesús De Loera and Luis Rademacher https://arxiv.org/abs/1710.02608 1 Minimum
More informationA Brief Look at Optimization
A Brief Look at Optimization CSC 412/2506 Tutorial David Madras January 18, 2018 Slides adapted from last year s version Overview Introduction Classes of optimization problems Linear programming Steepest
More informationLecture 9: Pipage Rounding Method
Recent Advances in Approximation Algorithms Spring 2015 Lecture 9: Pipage Rounding Method Lecturer: Shayan Oveis Gharan April 27th Disclaimer: These notes have not been subjected to the usual scrutiny
More informationJessica Su (some parts copied from CLRS / last quarter s notes)
1 Max flow Consider a directed graph G with positive edge weights c that define the capacity of each edge. We can identify two special nodes in G: the source node s and the sink node t. We want to find
More informationCombinatorial optimization - Structures and Algorithms, GeorgiaTech, Fall 2011 Lectures 1 4: Aug 23 Sep 1
Combinatorial optimization - Structures and Algorithms, GeorgiaTech, Fall 011 Lectures 1 4: Aug 3 Sep 1 László Végh References are from the book Connections in Combinatorial Optimization by András Frank
More informationIncremental Gradient, Subgradient, and Proximal Methods for Convex Optimization: A Survey. Chapter 4 : Optimization for Machine Learning
Incremental Gradient, Subgradient, and Proximal Methods for Convex Optimization: A Survey Chapter 4 : Optimization for Machine Learning Summary of Chapter 2 Chapter 2: Convex Optimization with Sparsity
More informationIE598 Big Data Optimization Summary Nonconvex Optimization
IE598 Big Data Optimization Summary Nonconvex Optimization Instructor: Niao He April 16, 2018 1 This Course Big Data Optimization Explore modern optimization theories, algorithms, and big data applications
More informationMS&E 213 / CS 269O : Chapter 1 Introduction to Introduction to Optimization Theory
MS&E 213 / CS 269O : Chapter 1 Introduction to Introduction to Optimization Theory By Aaron Sidford (sidford@stanford.edu) April 29, 2017 1 What is this course about? The central problem we consider throughout
More informationImportant separators and parameterized algorithms
Important separators and parameterized algorithms Dániel Marx 1 1 Institute for Computer Science and Control, Hungarian Academy of Sciences (MTA SZTAKI) Budapest, Hungary School on Parameterized Algorithms
More informationCombinatorial meets Convex Leveraging combinatorial structure to obtain faster algorithms, provably
Combinatorial meets Convex Leveraging combinatorial structure to obtain faster algorithms, provably Lorenzo Orecchia Department of Computer Science 6/30/15 Challenges in Optimization for Data Science 1
More information/ Approximation Algorithms Lecturer: Michael Dinitz Topic: Linear Programming Date: 2/24/15 Scribe: Runze Tang
600.469 / 600.669 Approximation Algorithms Lecturer: Michael Dinitz Topic: Linear Programming Date: 2/24/15 Scribe: Runze Tang 9.1 Linear Programming Suppose we are trying to approximate a minimization
More informationDavid G. Luenberger Yinyu Ye. Linear and Nonlinear. Programming. Fourth Edition. ö Springer
David G. Luenberger Yinyu Ye Linear and Nonlinear Programming Fourth Edition ö Springer Contents 1 Introduction 1 1.1 Optimization 1 1.2 Types of Problems 2 1.3 Size of Problems 5 1.4 Iterative Algorithms
More informationIE521 Convex Optimization Introduction
IE521 Convex Optimization Introduction Instructor: Niao He Jan 18, 2017 1 About Me Assistant Professor, UIUC, 2016 Ph.D. in Operations Research, M.S. in Computational Sci. & Eng. Georgia Tech, 2010 2015
More informationAlternating Direction Method of Multipliers
Alternating Direction Method of Multipliers CS 584: Big Data Analytics Material adapted from Stephen Boyd (https://web.stanford.edu/~boyd/papers/pdf/admm_slides.pdf) & Ryan Tibshirani (http://stat.cmu.edu/~ryantibs/convexopt/lectures/21-dual-meth.pdf)
More informationAn Introduction to LP Relaxations for MAP Inference
An Introduction to LP Relaxations for MAP Inference Adrian Weller MLSALT4 Lecture Feb 27, 2017 With thanks to David Sontag (NYU) for use of some of his slides and illustrations For more information, see
More informationDynamic Planar-Cuts: Efficient Computation of Min-Marginals for Outer-Planar Models
Dynamic Planar-Cuts: Efficient Computation of Min-Marginals for Outer-Planar Models Dhruv Batra 1 1 Carnegie Mellon University www.ece.cmu.edu/ dbatra Tsuhan Chen 2,1 2 Cornell University tsuhan@ece.cornell.edu
More informationMinimum-Cost Flow Network
Minimum-Cost Flow Network Smoothed flow network: G = (V,E) balance values: b : V Z costs: c : E R capacities: u : E N 2 1/2 3/1-1 1/3 3/2 3/1 1 3/1 1/2-2 cost/capacity 2/ 17 Smoothed 2 1 1/2 3/1 1 1 11/3
More informationLecture 10,11: General Matching Polytope, Maximum Flow. 1 Perfect Matching and Matching Polytope on General Graphs
CMPUT 675: Topics in Algorithms and Combinatorial Optimization (Fall 2009) Lecture 10,11: General Matching Polytope, Maximum Flow Lecturer: Mohammad R Salavatipour Date: Oct 6 and 8, 2009 Scriber: Mohammad
More informationGraph Cuts. Srikumar Ramalingam School of Computing University of Utah
Graph Cuts Srikumar Ramalingam School o Computing University o Utah Outline Introduction Pseudo-Boolean Functions Submodularity Max-low / Min-cut Algorithm Alpha-Expansion Segmentation Problem [Boykov
More information5 Machine Learning Abstractions and Numerical Optimization
Machine Learning Abstractions and Numerical Optimization 25 5 Machine Learning Abstractions and Numerical Optimization ML ABSTRACTIONS [some meta comments on machine learning] [When you write a large computer
More informationBilinear Programming
Bilinear Programming Artyom G. Nahapetyan Center for Applied Optimization Industrial and Systems Engineering Department University of Florida Gainesville, Florida 32611-6595 Email address: artyom@ufl.edu
More informationModels for grids. Computer vision: models, learning and inference. Multi label Denoising. Binary Denoising. Denoising Goal.
Models for grids Computer vision: models, learning and inference Chapter 9 Graphical Models Consider models where one unknown world state at each pixel in the image takes the form of a grid. Loops in the
More information1 Overview. 2 Applications of submodular maximization. AM 221: Advanced Optimization Spring 2016
AM : Advanced Optimization Spring 06 Prof. Yaron Singer Lecture 0 April th Overview Last time we saw the problem of Combinatorial Auctions and framed it as a submodular maximization problem under a partition
More informationProbabilistic Graphical Models
School of Computer Science Probabilistic Graphical Models Theory of Variational Inference: Inner and Outer Approximation Eric Xing Lecture 14, February 29, 2016 Reading: W & J Book Chapters Eric Xing @
More informationDistributed Submodular Maximization in Massive Datasets. Alina Ene. Joint work with Rafael Barbosa, Huy L. Nguyen, Justin Ward
Distributed Submodular Maximization in Massive Datasets Alina Ene Joint work with Rafael Barbosa, Huy L. Nguyen, Justin Ward Combinatorial Optimization Given A set of objects V A function f on subsets
More informationCS 435, 2018 Lecture 2, Date: 1 March 2018 Instructor: Nisheeth Vishnoi. Convex Programming and Efficiency
CS 435, 2018 Lecture 2, Date: 1 March 2018 Instructor: Nisheeth Vishnoi Convex Programming and Efficiency In this lecture, we formalize convex programming problem, discuss what it means to solve it efficiently
More informationConvex Optimization / Homework 2, due Oct 3
Convex Optimization 0-725/36-725 Homework 2, due Oct 3 Instructions: You must complete Problems 3 and either Problem 4 or Problem 5 (your choice between the two) When you submit the homework, upload a
More informationOptimization of submodular functions
Optimization of submodular functions Law Of Diminishing Returns According to the Law of Diminishing Returns (also known as Diminishing Returns Phenomenon), the value or enjoyment we get from something
More informationarxiv: v1 [cs.cv] 5 Mar 2015
Convex Optimization for Parallel Energy Minimization arxiv:1503.01563v1 [cs.cv] 5 Mar 2015 K. S. Sesh Kumar INRIA-Sierra project-team Département d Informatique de l Ecole Normale Supérieure Paris, France
More informationDS 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 informationLinear Regression and K-Nearest Neighbors 3/28/18
Linear Regression and K-Nearest Neighbors 3/28/18 Linear Regression Hypothesis Space Supervised learning For every input in the data set, we know the output Regression Outputs are continuous A number,
More informationIntroduction à la vision artificielle X
Introduction à la vision artificielle X Jean Ponce Email: ponce@di.ens.fr Web: http://www.di.ens.fr/~ponce Planches après les cours sur : http://www.di.ens.fr/~ponce/introvis/lect10.pptx http://www.di.ens.fr/~ponce/introvis/lect10.pdf
More informationSelected Topics in Column Generation
Selected Topics in Column Generation February 1, 2007 Choosing a solver for the Master Solve in the dual space(kelly s method) by applying a cutting plane algorithm In the bundle method(lemarechal), a
More informationGraph Cuts. Srikumar Ramalingam School of Computing University of Utah
Graph Cuts Srikumar Ramalingam School o Computing University o Utah Outline Introduction Pseudo-Boolean Functions Submodularity Max-low / Min-cut Algorithm Alpha-Expansion Segmentation Problem [Boykov
More informationAnalyzing Stochastic Gradient Descent for Some Non- Convex Problems
Analyzing Stochastic Gradient Descent for Some Non- Convex Problems Christopher De Sa Soon at Cornell University cdesa@stanford.edu stanford.edu/~cdesa Kunle Olukotun Christopher Ré Stanford University
More informationMonotone Closure of Relaxed Constraints in Submodular Optimization: Connections Between Minimization and Maximization: Extended Version
Monotone Closure of Relaxed Constraints in Submodular Optimization: Connections Between Minimization and Maximization: Extended Version Rishabh Iyer Dept. of Electrical Engineering University of Washington
More informationLarge Scale Density-friendly Graph Decomposition via Convex Programming
Large Scale Density-friendly Graph Decomposition via Convex Programming Maximilien Danisch LTCI, Télécom ParisTech, Université Paris-Saclay, 753, Paris, France danisch@telecomparistech.fr T-H. Hubert Chan
More informationRegularization and Markov Random Fields (MRF) CS 664 Spring 2008
Regularization and Markov Random Fields (MRF) CS 664 Spring 2008 Regularization in Low Level Vision Low level vision problems concerned with estimating some quantity at each pixel Visual motion (u(x,y),v(x,y))
More informationConditional gradient algorithms for machine learning
Conditional gradient algorithms for machine learning Zaid Harchaoui LEAR-LJK, INRIA Grenoble Anatoli Juditsky LJK, Université de Grenoble Arkadi Nemirovski Georgia Tech Abstract We consider penalized formulations
More informationSpicyMKL Efficient multiple kernel learning method using dual augmented Lagrangian
SpicyMKL Efficient multiple kernel learning method using dual augmented Lagrangian Taiji Suzuki Ryota Tomioka The University of Tokyo Graduate School of Information Science and Technology Department of
More informationRecent Advances in Frank-Wolfe Optimization. Simon Lacoste-Julien
Recent Advances in Frank-Wolfe Optimization Simon Lacoste-Julien OSL 2017 Les Houches April 13 th, 2017 Outline Frank-Wolfe algorithm review global linear convergence of FW optimization variants condition
More informationNetwork Flows. 1. Flows deal with network models where edges have capacity constraints. 2. Shortest paths deal with edge costs.
Network Flows. Flows deal with network models where edges have capacity constraints. 2. Shortest paths deal with edge costs. 4 u v 2 5 s 6 5 t 4 x 2 y 3 3. Network flow formulation: A network G = (V, E).
More informationDC Programming: A brief tutorial. Andres Munoz Medina Courant Institute of Mathematical Sciences
DC Programming: A brief tutorial. Andres Munoz Medina Courant Institute of Mathematical Sciences munoz@cims.nyu.edu Difference of Convex Functions Definition: A function exists f is said to be DC if there
More informationCS675: Convex and Combinatorial Optimization Spring 2018 The Simplex Algorithm. Instructor: Shaddin Dughmi
CS675: Convex and Combinatorial Optimization Spring 2018 The Simplex Algorithm Instructor: Shaddin Dughmi Algorithms for Convex Optimization We will look at 2 algorithms in detail: Simplex and Ellipsoid.
More informations-t Graph Cuts for Binary Energy Minimization
s-t Graph Cuts for Binary Energy Minimization E data term ( L) = D ( ) + ( ) P Lp λ Ι Lp Lq p prior term Now that we have an energy function, the big question is how do we minimize it? pq N n Exhaustive
More informationOutline of this Talk
Outline of this Talk Data Association associate common detections across frames matching up who is who Two frames: linear assignment problem Generalize to three or more frames increasing solution quality
More informationLearning Low-order Models for Enforcing High-order Statistics
Patrick Pletscher ETH Zurich Zurich, Switzerland Pushmeet Kohli Microsoft Research Cambridge, UK Abstract Models such as pairwise conditional random fields (CRFs) are extremely popular in computer vision
More informationCS261: Problem Set #1
CS261: Problem Set #1 Due by 11:59 PM on Tuesday, April 21, 2015 Instructions: (1) Form a group of 1-3 students. You should turn in only one write-up for your entire group. (2) Turn in your solutions by
More informationMVE165/MMG630, Applied Optimization Lecture 8 Integer linear programming algorithms. Ann-Brith Strömberg
MVE165/MMG630, Integer linear programming algorithms Ann-Brith Strömberg 2009 04 15 Methods for ILP: Overview (Ch. 14.1) Enumeration Implicit enumeration: Branch and bound Relaxations Decomposition methods:
More informationLagrangian Relaxation: An overview
Discrete Math for Bioinformatics WS 11/12:, by A. Bockmayr/K. Reinert, 22. Januar 2013, 13:27 4001 Lagrangian Relaxation: An overview Sources for this lecture: D. Bertsimas and J. Tsitsiklis: Introduction
More informationDivide and Conquer Kernel Ridge Regression
Divide and Conquer Kernel Ridge Regression Yuchen Zhang John Duchi Martin Wainwright University of California, Berkeley COLT 2013 Yuchen Zhang (UC Berkeley) Divide and Conquer KRR COLT 2013 1 / 15 Problem
More informationMarkov/Conditional Random Fields, Graph Cut, and applications in Computer Vision
Markov/Conditional Random Fields, Graph Cut, and applications in Computer Vision Fuxin Li Slides and materials from Le Song, Tucker Hermans, Pawan Kumar, Carsten Rother, Peter Orchard, and others Recap:
More informationIntroduction Optimization Geoff Gordon Ryan Tibshirani
Introduction 10-75 Optimization Geoff Gordon Ryan Tibshirani Administrivia http://www.cs.cmu.edu/~ggordon/1075-f1/ http://groups.google.com/group/1075-f1 Administrivia Prerequisites: no formal ones, but
More informationWeek 5. Convex Optimization
Week 5. Convex Optimization Lecturer: Prof. Santosh Vempala Scribe: Xin Wang, Zihao Li Feb. 9 and, 206 Week 5. Convex Optimization. The convex optimization formulation A general optimization problem is
More informationAspects of Convex, Nonconvex, and Geometric Optimization (Lecture 1) Suvrit Sra Massachusetts Institute of Technology
Aspects of Convex, Nonconvex, and Geometric Optimization (Lecture 1) Suvrit Sra Massachusetts Institute of Technology Hausdorff Institute for Mathematics (HIM) Trimester: Mathematics of Signal Processing
More informationDETERMINISTIC OPERATIONS RESEARCH
DETERMINISTIC OPERATIONS RESEARCH Models and Methods in Optimization Linear DAVID J. RADER, JR. Rose-Hulman Institute of Technology Department of Mathematics Terre Haute, IN WILEY A JOHN WILEY & SONS,
More informationNotes on Minimum Cuts and Modular Functions
Notes on Minimum Cuts and Modular Functions 1 Introduction The following are my notes on Cunningham s paper [1]. Given a submodular function f and a set S, submodular minimisation is the problem of finding
More information1. Introduction. performance of numerical methods. complexity bounds. structural convex optimization. course goals and topics
1. Introduction EE 546, Univ of Washington, Spring 2016 performance of numerical methods complexity bounds structural convex optimization course goals and topics 1 1 Some course info Welcome to EE 546!
More informationChapter 5 Graph Algorithms Algorithm Theory WS 2012/13 Fabian Kuhn
Chapter 5 Graph Algorithms Algorithm Theory WS 2012/13 Fabian Kuhn Graphs Extremely important concept in computer science Graph, : node (or vertex) set : edge set Simple graph: no self loops, no multiple
More informationData Mining Chapter 8: Search and Optimization Methods Fall 2011 Ming Li Department of Computer Science and Technology Nanjing University
Data Mining Chapter 8: Search and Optimization Methods Fall 2011 Ming Li Department of Computer Science and Technology Nanjing University Search & Optimization Search and Optimization method deals with
More informationNatural Language Processing
Natural Language Processing Classification III Dan Klein UC Berkeley 1 Classification 2 Linear Models: Perceptron The perceptron algorithm Iteratively processes the training set, reacting to training errors
More informationLecture 12: Feasible direction methods
Lecture 12 Lecture 12: Feasible direction methods Kin Cheong Sou December 2, 2013 TMA947 Lecture 12 Lecture 12: Feasible direction methods 1 / 1 Feasible-direction methods, I Intro Consider the problem
More informationColumn-Action Methods in Image Reconstruction
Column-Action Methods in Image Reconstruction Per Christian Hansen joint work with Tommy Elfving Touraj Nikazad Overview of Talk Part 1: the classical row-action method = ART The advantage of algebraic
More informationMaximum flows & Maximum Matchings
Chapter 9 Maximum flows & Maximum Matchings This chapter analyzes flows and matchings. We will define flows and maximum flows and present an algorithm that solves the maximum flow problem. Then matchings
More information/633 Introduction to Algorithms Lecturer: Michael Dinitz Topic: Approximation algorithms Date: 11/27/18
601.433/633 Introduction to Algorithms Lecturer: Michael Dinitz Topic: Approximation algorithms Date: 11/27/18 22.1 Introduction We spent the last two lectures proving that for certain problems, we can
More informationSegmentation with non-linear constraints on appearance, complexity, and geometry
IPAM February 2013 Western Univesity Segmentation with non-linear constraints on appearance, complexity, and geometry Yuri Boykov Andrew Delong Lena Gorelick Hossam Isack Anton Osokin Frank Schmidt Olga
More informationCS 473: Algorithms. Ruta Mehta. Spring University of Illinois, Urbana-Champaign. Ruta (UIUC) CS473 1 Spring / 36
CS 473: Algorithms Ruta Mehta University of Illinois, Urbana-Champaign Spring 2018 Ruta (UIUC) CS473 1 Spring 2018 1 / 36 CS 473: Algorithms, Spring 2018 LP Duality Lecture 20 April 3, 2018 Some of the
More informationPlanar Cycle Covering Graphs
Planar Cycle Covering Graphs Julian Yarkony, Alexander T. Ihler, Charless C. Fowlkes Department of Computer Science University of California, Irvine {jyarkony,ihler,fowlkes}@ics.uci.edu Abstract We describe
More informationSpeeding up MWU based Approximation Schemes and Some Applications
Speeding up MWU based Approximation Schemes and Some Applications Chandra Chekuri Univ. of Illinois, Urbana-Champaign Joint work with Kent Quanrud Based on recent papers Near-linear-time approximation
More informationAn Interface-fitted Mesh Generator and Polytopal Element Methods for Elliptic Interface Problems
An Interface-fitted Mesh Generator and Polytopal Element Methods for Elliptic Interface Problems Long Chen University of California, Irvine chenlong@math.uci.edu Joint work with: Huayi Wei (Xiangtan University),
More informationEnergy Minimization Under Constraints on Label Counts
Energy Minimization Under Constraints on Label Counts Yongsub Lim 1, Kyomin Jung 1, and Pushmeet Kohli 2 1 Korea Advanced Istitute of Science and Technology, Daejeon, Korea yongsub@kaist.ac.kr, kyomin@kaist.edu
More informationSubmodularity Cuts and Applications
Submodularity Cuts and Applications Yoshinobu Kawahara The Inst. of Scientific and Industrial Res. (ISIR), Osaka Univ., Japan kawahara@ar.sanken.osaka-u.ac.jp Koji Tsuda Comp. Bio. Research Center, AIST,
More informationDiscrete Optimization of Ray Potentials for Semantic 3D Reconstruction
Discrete Optimization of Ray Potentials for Semantic 3D Reconstruction Marc Pollefeys Joined work with Nikolay Savinov, Christian Haene, Lubor Ladicky 2 Comparison to Volumetric Fusion Higher-order ray
More informationSubmodularity Reading Group. Matroid Polytopes, Polymatroid. M. Pawan Kumar
Submodularity Reading Group Matroid Polytopes, Polymatroid M. Pawan Kumar http://www.robots.ox.ac.uk/~oval/ Outline Linear Programming Matroid Polytopes Polymatroid Polyhedron Ax b A : m x n matrix b:
More informationLecture 2 - Introduction to Polytopes
Lecture 2 - Introduction to Polytopes Optimization and Approximation - ENS M1 Nicolas Bousquet 1 Reminder of Linear Algebra definitions Let x 1,..., x m be points in R n and λ 1,..., λ m be real numbers.
More informationStructured 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 informationDS 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 January 24 2019 Logistics HW 1 is due on Friday 01/25 Project proposal: due Feb 21 1 page description
More informationSequential and parallel deficit scaling algorithms for minimum flow in bipartite networks
Sequential and parallel deficit scaling algorithms for minimum flow in bipartite networks LAURA CIUPALĂ Department of Computer Science University Transilvania of Braşov Iuliu Maniu Street 50, Braşov ROMANIA
More informationMemory Bandwidth and Low Precision Computation. CS6787 Lecture 9 Fall 2017
Memory Bandwidth and Low Precision Computation CS6787 Lecture 9 Fall 2017 Memory as a Bottleneck So far, we ve just been talking about compute e.g. techniques to decrease the amount of compute by decreasing
More informationInvestigating Mixed-Integer Hulls using a MIP-Solver
Investigating Mixed-Integer Hulls using a MIP-Solver Matthias Walter Otto-von-Guericke Universität Magdeburg Joint work with Volker Kaibel (OvGU) Aussois Combinatorial Optimization Workshop 2015 Outline
More informationIntegral Geometry and the Polynomial Hirsch Conjecture
Integral Geometry and the Polynomial Hirsch Conjecture Jonathan Kelner, MIT Partially based on joint work with Daniel Spielman Introduction n A lot of recent work on Polynomial Hirsch Conjecture has focused
More informationECS289: 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 informationCS 435, 2018 Lecture 9, Date: 3 May 2018 Instructor: Nisheeth Vishnoi. Cutting Plane and Ellipsoid Methods for Linear Programming
CS 435, 2018 Lecture 9, Date: 3 May 2018 Instructor: Nisheeth Vishnoi Cutting Plane and Ellipsoid Methods for Linear Programming In this lecture we introduce the class of cutting plane methods for convex
More informationStructured Optimal Transport
David Alvarez-Melis Tommi S. Jaakkola Stefanie Jegelka MIT CSAIL MIT CSAIL MIT CSAIL Abstract Optimal Transport has recently gained interest in machine learning for applications ranging from domain adaptation
More informationCPSC 340: Machine Learning and Data Mining. Principal Component Analysis Fall 2017
CPSC 340: Machine Learning and Data Mining Principal Component Analysis Fall 2017 Assignment 3: 2 late days to hand in tonight. Admin Assignment 4: Due Friday of next week. Last Time: MAP Estimation MAP
More information2 The Mixed Postman Problem with Restrictions on the Arcs
Approximation Algorithms for the Mixed Postman Problem with Restrictions on the Arcs Francisco Javier Zaragoza Martínez Departamento de Sistemas, Universidad Autónoma Metropolitana Unidad Azcapotzalco
More informationOn Distributed Submodular Maximization with Limited Information
On Distributed Submodular Maximization with Limited Information Bahman Gharesifard Stephen L. Smith Abstract This paper considers a class of distributed submodular maximization problems in which each agent
More informationThe convex geometry of inverse problems
The convex geometry of inverse problems Benjamin Recht Department of Computer Sciences University of Wisconsin-Madison Joint work with Venkat Chandrasekaran Pablo Parrilo Alan Willsky Linear Inverse Problems
More informationOnline Submodular Minimization for Combinatorial Structures
Stefanie Jegelka Max Planck Institute for Intelligent Systems, Tübingen, Germany Jeff Bilmes University of Washington, Seattle, WA 98195, USA jegelka@tuebingen.mpg.de bilmes@u.washington.edu Abstract Most
More informationSparse Optimization Lecture: Proximal Operator/Algorithm and Lagrange Dual
Sparse Optimization Lecture: Proximal Operator/Algorithm and Lagrange Dual Instructor: Wotao Yin July 2013 online discussions on piazza.com Those who complete this lecture will know learn the proximal
More informationarxiv: v1 [cs.ds] 19 Nov 2018
Towards Nearly-linear Time Algorithms for Submodular Maximization with a Matroid Constraint Alina Ene Huy L. Nguy ên arxiv:1811.07464v1 [cs.ds] 19 Nov 2018 Abstract We consider fast algorithms for monotone
More informationThe Simplex Algorithm
The Simplex Algorithm Uri Feige November 2011 1 The simplex algorithm The simplex algorithm was designed by Danzig in 1947. This write-up presents the main ideas involved. It is a slight update (mostly
More informationCS6670: Computer Vision
CS6670: Computer Vision Noah Snavely Lecture 19: Graph Cuts source S sink T Readings Szeliski, Chapter 11.2 11.5 Stereo results with window search problems in areas of uniform texture Window-based matching
More informationCommunication Complexity of Combinatorial Auctions with Submodular Valuations
Communication Complexity of Combinatorial Auctions with Submodular Valuations Shahar Dobzinski Weizmann Institute of Science Jan Vondrák IBM Almaden Research ACM-SIAM SODA 2013, New Orleans Dobzinski-Vondrák
More informationWhat have we leaned so far?
What have we leaned so far? Camera structure Eye structure Project 1: High Dynamic Range Imaging What have we learned so far? Image Filtering Image Warping Camera Projection Model Project 2: Panoramic
More information