Learning to Learn with Compound HD Models

Ruslan Salakhutdinov, Joshua B. Tenenbaum, Antonio Torralba
2011 Neural Information Processing Systems  
We introduce HD (or "Hierarchical-Deep") models, a new compositional learning architecture that integrates deep learning models with structured hierarchical Bayesian 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 examples, by learning low-level generic features, high-level features that
more » ... re 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.
dblp:conf/nips/SalakhutdinovTT11 fatcat:zgmboicurrcntkx2hih33ylhmm