《Learning with Hierarchical-Deep Models》.pdf

1958 IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 35, NO. 8, AUGUST 2013 Learning with Hierarchical-Deep Models Ruslan Salakhutdinov, Joshua B. Tenenbaum, and Antonio Torralba, Member, IEEE Abstract—We introduce HD (or “Hierarchical-Deep”) models, a new compositional learning architecture that integrates deep learning models with structured hierarchical Bayesian (HB) models. Specifically, we show how we can learn a hierarchical Dirichlet process (HDP) prior over the activities of the top-level features in a deep Boltzmann machine (DBM). This compound HDP-DBM model learns to learn novel concepts from very few training example by learning low-level generic features, high-level features that capture correlations among low-level features, and a category hierarchy for sharing priors over the high-level features that are typical of different kinds of concepts. We present efficient learning and inference algorithms for the HDP-DBM model and show that it is able to learn new concepts from very few examples on CIFAR-100 object recognition, handwritten character recognition, and human motion capture datasets. Index Terms—Deep networks, deep Boltzmann machines, hierarchical Bayesian models, one-shot learning Ç 1 INTRODUCTION THE ability to learn abstract representations that support representations for many high-dimensional datasets. The transfer to novel but related tasks lies at the core of ability to automatically learn in multiple layers allows deep many problems in computer vision, natural language models to construct sophisticated domain-specific features processing, cognitive science, and machine learning. In wi