This paper proposes a general learning framework for a class of problems that require learning over latent intermediate representations. Many natural language processing (NLP) dec...
Ming-Wei Chang, Dan Goldwasser, Dan Roth, Vivek Sr...
One approach to modeling structured discrete data is to describe the probability of states via an energy function and Gibbs distribution. A recurring difficulty in these models is...
Daniel Tarlow, Ryan Prescott Adams, Richard S. Zem...
Predictive State Representations (PSRs) have shown a great deal of promise as an alternative to Markov models. However, learning a PSR from a single stream of data generated from ...
Training principles for unsupervised learning are often derived from motivations that appear to be independent of supervised learning. In this paper we present a simple unificatio...
We provide a novel view of learning an approximate model of a partially observable environment from data and present a simple implemenf the idea. The learned model abstracts away ...