TTOpt: A Maximum Volume Quantized Tensor Train-Based Optimization and Its Application to Reinforcement Learning
Abstract
We present a novel procedure for optimization based on the combination of efficient quantized tensor train representation and a generalized maximum matrix volume principle.We demonstrate the applicability of the new Tensor Train Optimizer (TTOpt) method for various tasks, ranging from minimization of multidimensional functions to reinforcement learning.Our algorithm compares favorably to popular gradient-free methods and outperforms them by the number of function evaluations or execution time, often by a significant margin.
Cite
Text
Sozykin et al. "TTOpt: A Maximum Volume Quantized Tensor Train-Based Optimization and Its Application to Reinforcement Learning." Neural Information Processing Systems, 2022.Markdown
[Sozykin et al. "TTOpt: A Maximum Volume Quantized Tensor Train-Based Optimization and Its Application to Reinforcement Learning." Neural Information Processing Systems, 2022.](https://mlanthology.org/neurips/2022/sozykin2022neurips-ttopt/)BibTeX
@inproceedings{sozykin2022neurips-ttopt,
title = {{TTOpt: A Maximum Volume Quantized Tensor Train-Based Optimization and Its Application to Reinforcement Learning}},
author = {Sozykin, Konstantin and Chertkov, Andrei and Schutski, Roman and Phan, Anh-Huy and Cichocki, Andrzej S and Oseledets, Ivan},
booktitle = {Neural Information Processing Systems},
year = {2022},
url = {https://mlanthology.org/neurips/2022/sozykin2022neurips-ttopt/}
}