LGApr 10, 2020

Efficient Sampled Softmax for Tensorflow

arXiv:2004.05244v11 citations
AI Analysis

This is an incremental improvement for Tensorflow users needing faster training with sampled softmax.

The paper tackled the problem of slow sampled softmax loss in Tensorflow by simplifying the graph for forward and backward passes, resulting in a speedup over the default implementation.

This short paper discusses an efficient implementation of \emph{sampled softmax loss} for Tensorflow. The speedup over the default implementation is achieved due to simplification of the graph for the forward and backward passes.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes