Integrated Inference and Learning of Neural Factors in Structural Support Vector Machines

Abstract

Tackling pattern recognition problems in areas such as computer vision, bioinformatics, speech or text recognition is often done best by taking into account task-specific statistical relations between output variables. In structured prediction, this internal structure is used to predict multiple outputs simultaneously, leading to more accurate and coherent predictions. Structural support vector machines (SSVMs) are nonprobabilistic models that optimize a joint input-output function through margin-based learning. Because SSVMs generally disregard the interplay between unary and interaction factors during the training phase, final parameters are suboptimal. Moreover, its factors are often restricted to linear combinations of input features, limiting its generalization power. To improve prediction accuracy, this paper proposes: (i) Joint inference and learning by integration of back-propagation and loss-augmented inference in SSVM subgradient descent; (ii) Extending SSVM factors to neural networks that form highly nonlinear functions of input features. Image segmentation benchmark results demonstrate improvements over conventional SSVM training methods in terms of accuracy, highlighting the feasibility of end-to-end SSVM training with neural factors.

References in corpus (13)

• Natural Language Processing (almost) from Scratch
• Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs
• Theano: new features and speed improvements
• Joint Training of a Convolutional Network and a Graphical Model for Human Pose Estimation
• Identifying and attacking the saddle point problem in high-dimensional non-convex optimization
• Deep Human Parsing with Active Template Regression
• Fully Connected Deep Structured Networks
• Block-Coordinate Frank-Wolfe Optimization for Structural SVMs
• Discriminatively Trained And-Or Graph Models for Object Shape Detection
• On the saddle point problem for non-convex optimization
• Incorporating Structural Alternatives and Sharing into Hierarchy for Multiclass Object Recognition and Detection
• Compositional Structure Learning for Action Understanding
• Structured Learning via Logistic Regression