Search Results for author: Takuya Inoue

Found 1 papers, 1 papers with code

A Graph Theoretic Framework of Recomputation Algorithms for Memory-Efficient Backpropagation

3 code implementations NeurIPS 2019 Mitsuru Kusumoto, Takuya Inoue, Gentaro Watanabe, Takuya Akiba, Masanori Koyama

Recomputation algorithms collectively refer to a family of methods that aims to reduce the memory consumption of the backpropagation by selectively discarding the intermediate results of the forward propagation and recomputing the discarded results as needed.

Cannot find the paper you are looking for? You can Submit a new open access paper.