Adversarial Learning for 3D Matching

Structured prediction of objects in spaces that are inherently difficult to search or compactly characterize is a particularly challenging task. For example, though bipartite matchings in two dimensions can be tractably optimized and learned, the higher-dimensional generalization—3D matchings—are NP-hard to optimally obtain and the set of potential solutions cannot be compactly characterized. Though approximation is therefore necessary, prevalent structured prediction methods inherit the weaknesses they possess in the twodimensional setting—either suffering from inconsistency or intractability—even when the approximations are sufficient. In this paper, we explore extending an adversarial approach to learning bipartite matchings that avoids these weaknesses to the three dimensional setting. We assess the benefits compared to marginbased methods on a three-frame tracking problem.

[1]  Viggo Kann,et al.  Maximum Bounded 3-Dimensional Matching is MAX SNP-Complete , 1991, Inf. Process. Lett..

[2]  Harold W. Kuhn,et al.  The Hungarian method for the assignment problem , 1955, 50 Years of Integer Programming.

[3]  Avrim Blum,et al.  Planning in the Presence of Cost Functions Controlled by an Adversary , 2003, ICML.

[4]  Thomas Hofmann,et al.  Large Margin Methods for Structured and Interdependent Output Variables , 2005, J. Mach. Learn. Res..

[5]  Ramakant Nevatia,et al.  Global data association for multi-object tracking using network flows , 2008, 2008 IEEE Conference on Computer Vision and Pattern Recognition.

[6]  Bernt Schiele,et al.  Subgraph decomposition for multi-target tracking , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[7]  I JordanMichael,et al.  Graphical Models, Exponential Families, and Variational Inference , 2008 .

[8]  Ben Taskar,et al.  Learning associative Markov networks , 2004, ICML.

[9]  Ivan Laptev,et al.  On pairwise costs for network flow multi-object tracking , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[10]  Thomas Hofmann,et al.  Support vector machine learning for interdependent and structured output spaces , 2004, ICML.

[11]  R. Fisher,et al.  On the Mathematical Foundations of Theoretical Statistics , 1922 .

[12]  Stephen P. Boyd,et al.  Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers , 2011, Found. Trends Mach. Learn..

[13]  A. Dawid,et al.  Game theory, maximum entropy, minimum discrepancy and robust Bayesian decision theory , 2004, math/0410076.

[14]  Brian D. Ziebart,et al.  Adversarial Cost-Sensitive Classification , 2015, UAI.

[15]  Jin Yu,et al.  Exponential Family Graph Matching and Ranking , 2009, NIPS.

[16]  Hong Wang,et al.  Adversarial Sequence Tagging , 2016, IJCAI.

[17]  Brian D. Ziebart,et al.  Adversarial Multiclass Classification: A Risk Minimization Perspective , 2016, NIPS.

[18]  Flemming Topsøe,et al.  Information-theoretical optimization techniques , 1979, Kybernetika.

[19]  M. Sion.  On general minimax theorems , 1958 .

[20]  John von Neumann,et al.  1. A Certain Zero-sum Two-person Game Equivalent to the Optimal Assignment Problem , 1953 .

[21]  Yoshua Bengio,et al.  Generative Adversarial Nets , 2014, NIPS.

[22]  Thomas Brox,et al.  A Multi-cut Formulation for Joint Segmentation and Tracking of Multiple Objects , 2016, ArXiv.

[23]  Marco Molinaro,et al.  Mixed-integer quadratic programming is in NP , 2014, Mathematical Programming.

[24]  Brian D. Ziebart,et al.  Adversarial Surrogate Losses for Ordinal Regression , 2017, NIPS.

[25]  E. Rowland.  Theory of Games and Economic Behavior , 1946, Nature.

[26]  H. H. Rachford,et al.  On the numerical solution of heat conduction problems in two and three space variables , 1956 .

[27]  Yoram Singer,et al.  Efficient projections onto the l1-ball for learning in high dimensions , 2008, ICML '08.

[28]  Brian D. Ziebart,et al.  Robust Classification Under Sample Selection Bias , 2014, NIPS.

[29]  Pedro M. Domingos,et al.  Adversarial classification , 2004, KDD.

[30]  Mark W. Schmidt,et al.  Optimizing Costly Functions with Simple Constraints: A Limited-Memory Projected Quasi-Newton Algorithm , 2009, AISTATS.

[31]  Bernt Schiele,et al.  Multiple People Tracking by Lifted Multicut and Person Re-identification , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[32]  Ben Taskar,et al.  Learning structured prediction models: a large margin approach , 2005, ICML.

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

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

[35]  Ambuj Tewari,et al.  On the Consistency of Multiclass Classification Methods , 2007, J. Mach. Learn. Res..

[36]  Zhu Han,et al.  Multi-block ADMM for big data optimization in smart grid , 2015, 2015 International Conference on Computing, Networking and Communications (ICNC).

[37]  Jan Feyereisl,et al.  Online Multi-target Tracking by Large Margin Structured Learning , 2012, ACCV.

[38]  Andrew McCallum,et al.  Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data , 2001, ICML.

[39]  Michael K. Ng,et al.  Birkhoff-von Neumann Theorem for Multistochastic Tensors , 2014, SIAM J. Matrix Anal. Appl..

[40]  D. Greig,et al.  Exact Maximum A Posteriori Estimation for Binary Images , 1989 .

[41]  Thorsten Joachims,et al.  A support vector method for multivariate performance measures , 2005, ICML.

[42]  Wei Xing,et al.  ARC: Adversarial Robust Cuts for Semi-Supervised and Multi-label Classification , 2018, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[43]  Xinhua Zhang,et al.  Efficient and Consistent Adversarial Bipartite Matching , 2018, ICML.

[44]  Fernando Pereira,et al.  Structured Learning with Approximate Inference , 2007, NIPS.