Inducing Value Sparsity for Parallel Inference in Tree-shaped Models

Abstract

Single-core architectures are rapidly on the decline, and even the most common computational devices now contain multiple cores. With this easy access to parallelism, the machine learning community needs to go beyond treating the running time as the only computational resource and needs to study approaches that take this additional form of flexibility into account. In this work, we study inference in tree-shaped models. Specifically, we focus on balancing accuracy and efficient use of multiple cores when faced with running time constraints.

Extracted Key Phrases

3 Figures and Tables

Cite this paper

@inproceedings{Singh2011InducingVS, title={Inducing Value Sparsity for Parallel Inference in Tree-shaped Models}, author={Sameer Singh and Brian Martin Andrew McCallum}, year={2011} }