Maximum persistency via iterative relaxed inference with graphical models

We consider the NP-hard problem of MAP-inference for graphical models. We propose a polynomial time practically efficient algorithm for finding a part of its optimal solution. Specifically, our algorithm marks each label in each node of the considered graphical model either as (i) optimal, meaning that it belongs to all optimal solutions of the inference problem; (ii) non-optimal if it provably does not belong to any solution; or (iii) undefined, which means our algorithm can not make a decision regarding the label. Moreover, we prove optimality of our approach: it delivers in a certain sense the largest total number of labels marked as optimal or non-optimal. We demonstrate superiority of our approach on problems from machine learning and computer vision benchmarks.

[1]  Nikos Komodakis,et al.  Approximate Labeling via Graph Cuts Based on Linear Programming , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[2]  Eugene C. Freuder.  Eliminating Interchangeable Values in Constraint Satisfaction Problems , 1991, AAAI.

[3]  Vladimir Kolmogorov,et al.  Convergent Tree-Reweighted Message Passing for Energy Minimization , 2006, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[4]  Olga Veksler,et al.  Fast approximate energy minimization via graph cuts , 2001, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[5]  Richard Szeliski,et al.  A Comparative Study of Energy Minimization Methods for Markov Random Fields with Smoothness-Based Priors , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  Vladimir Kolmogorov,et al.  Optimizing Binary MRFs via Extended Roof Duality , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[7]  Stanislav Zivny,et al.  The Complexity of Finite-Valued CSPs , 2016, J. ACM.

[8]  Robert J. Vanderbei,et al.  Linear Programming: Foundations and Extensions , 1998, Kluwer international series in operations research and management service.

[9]  Martin C. Cooper,et al.  Soft arc consistency revisited , 2010, Artif. Intell..

[10]  Joseph Naor,et al.  Approximation algorithms for the metric labeling problem via a new linear programming formulation , 2001, SODA '01.

[11]  Alok Aggarwal,et al.  Geometric applications of a matrix-searching algorithm , 1987, SCG '86.

[12]  Lise Getoor,et al.  Unifying Local Consistency and MAX SAT Relaxations for Scalable Inference with Rounding Guarantees , 2015, AISTATS.

[13]  Hanif D. Sherali,et al.  A Hierarchy of Relaxations Between the Continuous and Convex Hull Representations for Zero-One Programming Problems , 1990, SIAM J. Discret. Math..

[14]  Endre Boros,et al.  Pseudo-Boolean optimization , 2002, Discret. Appl. Math..

[15]  Leslie E. Trotter,et al.  Vertex packings: Structural properties and algorithms , 1975, Math. Program..

[16]  Martin J. Wainwright,et al.  Exact MAP Estimates by (Hyper)tree Agreement , 2002, NIPS.

[17]  Arie M. C. A. Koster,et al.  The partial constraint satisfaction problem: Facets and lifting theorems , 1998, Oper. Res. Lett..

[18]  Sebastian Nowozin,et al.  A Comparative Study of Modern Inference Techniques for Structured Discrete Energy Minimization Problems , 2014, International Journal of Computer Vision.

[19]  Chen Wang,et al.  Relaxation-Based Preprocessing Techniques for Markov Random Field Inference , 2016, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[20]  Christopher Zach,et al.  A Principled Approach for Coarse-to-Fine MAP Inference , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[21]  Bogdan Savchynskyy,et al.  Getting Feasible Variable Estimates from Infeasible Ones: MRF Local Polytope Study , 2012, 2013 IEEE International Conference on Computer Vision Workshops.

[22]  Barry O'Sullivan,et al.  Dead-End Elimination for Weighted CSP , 2013, CP.

[23]  Giorgio Gambosi,et al.  Complexity and approximation: combinatorial optimization problems and their approximability properties , 1999 .

[24]  A. Shekhovtsov.  Exact and Partial Energy Minimization in Computer Vision , 2013 .

[25]  Hiroshi Ishikawa,et al.  Transformation of General Binary MRF Minimization to the First-Order Case , 2011, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[26]  Daniel Cremers,et al.  Generalized Roof Duality for Multi-Label Optimization: Optimal Lower Bounds and Persistency , 2012, ECCV.

[27]  Endre Boros,et al.  A graph cut algorithm for higher-order Markov Random Fields , 2011, 2011 International Conference on Computer Vision.

[28]  Yair Weiss,et al.  Minimizing and Learning Energy Functions for Side-Chain Prediction , 2007, RECOMB.

[29]  Michael I. Jordan,et al.  Graphical Models, Exponential Families, and Variational Inference , 2008, Found. Trends Mach. Learn..

[30]  Éva Tardos,et al.  Approximation algorithms for classification problems with pairwise relationships: metric labeling and Markov random fields , 2002, JACM.

[31]  M. I. Schlesingera,et al.  Diffusion algorithms and structural recognition optimization problems , 2011 .

[32]  P. Orponen,et al.  On Approximation Preserving Reductions: Complete Problems and Robust Measures (Revised Version) , 1987 .

[33]  Tomás Werner,et al.  Universality of the Local Marginal Polytope , 2013, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[34]  Vladimir Kolmogorov,et al.  What energy functions can be minimized via graph cuts? , 2002, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[35]  Tomio Hirata,et al.  A Unified Linear-Time Algorithm for Computing Distance Maps , 1996, Inf. Process. Lett..

[36]  Tommi S. Jaakkola,et al.  Fixing Max-Product: Convergent Message Passing Algorithms for MAP LP-Relaxations , 2007, NIPS.

[37]  J. C. Picard,et al.  On the integer-valued variables in the linear vertex packing problem , 1977, Math. Program..

[38]  Rob A. Rutenbar,et al.  Fast hierarchical implementation of sequential tree-reweighted belief propagation for probabilistic inference , 2015, 2015 25th International Conference on Field Programmable Logic and Applications (FPL).

[39]  Pierre Hansen,et al.  Roof duality, complementation and persistency in quadratic 0–1 optimization , 1984, Math. Program..

[40]  Vladimir Kolmogorov,et al.  Generalized roof duality and bisubmodular functions , 2010, Discret. Appl. Math..

[41]  Wim H. Hesselink,et al.  A General Algorithm for Computing Distance Transforms in Linear Time , 2000, ISMM.

[42]  Christoph Schnörr,et al.  Partial Optimality by Pruning for MAP-Inference with General Graphical Models , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[43]  Vladimir Kolmogorov,et al.  The Power of Linear Programming for Finite-Valued CSPs: A Constructive Characterization , 2013, ICALP.

[44]  M. I. Schlesinger,et al.  Some solvable subclasses of structural recognition problems , 2000 .

[45]  Vladimir Kolmogorov,et al.  A New Look at Reweighted Message Passing , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[46]  Vladimir Kolmogorov,et al.  Potts Model, Parametric Maxflow and K-Submodular Functions , 2013, 2013 IEEE International Conference on Computer Vision.

[47]  Christoph Schnörr,et al.  A study of Nesterov's scheme for Lagrangian decomposition and MAP labeling , 2011, CVPR 2011.

[48]  Rob A. Rutenbar,et al.  Hardware implementation of MRF map inference on an FPGA platform , 2012, 22nd International Conference on Field Programmable Logic and Applications (FPL).

[49]  Christoph Schnörr,et al.  Efficient MRF Energy Minimization via Adaptive Diminishing Smoothing , 2012, UAI.

[50]  Johan Desmet,et al.  The dead-end elimination theorem and its use in protein side-chain positioning , 1992, Nature.

[51]  M. Shlezinger.  Syntactic analysis of two-dimensional visual signals in the presence of noise , 1976 .

[52]  Pushmeet Kohli,et al.  Reduce, reuse & recycle: Efficiently solving multi-label MRFs , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[53]  Ashraf M. Abdelbar,et al.  Approximating MAPs for Belief Networks is NP-Hard and Other Theorems , 1998, Artif. Intell..

[54]  Ivan Kovtun,et al.  Partial Optimal Labeling Search for a NP-Hard Subclass of (max, +) Problems , 2003, DAGM-Symposium.

[55]  Olivier Roussel,et al.  WCSP Integration of Soft Neighborhood Substitutability , 2012, CP.

[56]  Thomas Pock,et al.  Solving Dense Image Matching in Real-Time using Discrete-Continuous Optimization , 2016, ArXiv.

[57]  Daniel Huber,et al.  Complexity of Discrete Energy Minimization Problems , 2016, ECCV.

[58]  Philip H. S. Torr,et al.  LP-relaxation of Binarized Energy Minimization ( Version 1 . 50 ) , 2008 .

[59]  Hanif D. Sherali,et al.  Persistency in 0-1 Polynomial Programming , 1998, Math. Oper. Res..

[60]  Sebastian Nowozin,et al.  A Comparative Study of Modern Inference Techniques for Discrete Energy Minimization Problems , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition.

[61]  Alexander Shekhovtsov,et al.  Maximum Persistency in Energy Minimization , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[62]  Michel Balinski,et al.  Integer Programming: Methods, Uses, Computations , 1965 .

[63]  R. Goldstein.  Efficient rotamer elimination applied to protein side-chains and related spin glasses. , 1994, Biophysical journal.

[64]  Vladimir Kolmogorov,et al.  On partial optimality in multi-label MRFs , 2008, ICML '08.

[65]  Fredrik Kahl,et al.  Generalized roof duality for pseudo-boolean optimization , 2011, 2011 International Conference on Computer Vision.

[66]  Alexander Shekhovtsov,et al.  Higher order maximum persistency and comparison theorems , 2015, Comput. Vis. Image Underst..

[67]  Tomás Werner,et al.  A Linear Programming Approach to Max-Sum Problem: A Review , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[68]  Vladimir Kolmogorov.  Reweighted message passing revisited , 2013, ArXiv.

[69]  Daniel P. Huttenlocher,et al.  Efficient Belief Propagation for Early Vision , 2004, Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2004. CVPR 2004..

[70]  Christoph Schnörr,et al.  Partial Optimality via Iterative Pruning for the Potts Model , 2013, SSVM.