evosax: Jax-based evolution strategies

RT Lange - Proceedings of the Companion Conference on Genetic …, 2023 - dl.acm.org
The deep learning revolution has greatly been accelerated by the'hardware lottery': Recent
advances in modern hardware accelerators, compilers and the availability of open-source …

QDax: A library for quality-diversity and population-based algorithms with hardware acceleration

F Chalumeau, B Lim, R Boige, M Allard… - Journal of Machine …, 2024 - jmlr.org
QDax is an open-source library with a streamlined and modular API for Quality-Diversity
(QD) optimisation algorithms in Jax. The library serves as a versatile tool for optimisation …

TorchRL: A data-driven decision-making library for PyTorch

A Bou, M Bettini, S Dittert, V Kumar, S Sodhani… - arXiv preprint arXiv …, 2023 - arxiv.org
PyTorch has ascended as a premier machine learning framework, yet it lacks a native and
comprehensive library for decision and control tasks suitable for large development teams …

Population-based reinforcement learning for combinatorial optimization

N Grinsztajn, D Furelos-Blanco, TD Barrett - arXiv preprint arXiv …, 2022 - arxiv.org
Applying reinforcement learning (RL) to combinatorial optimization problems is attractive as
it removes the need for expert knowledge or pre-solved instances. However, it is unrealistic …

Going faster to see further: GPU-accelerated value iteration and simulation for perishable inventory control using JAX

J Farrington, K Li, WK Wong, M Utley - arXiv preprint arXiv:2303.10672, 2023 - arxiv.org
Value iteration can find the optimal replenishment policy for a perishable inventory problem,
but is computationally demanding due to the large state spaces that are required to …

Population-Based Reinforcement Learning for Combinatorial Optimization Problems

N Grinsztajn, D Furelos-Blanco, TD Barrett - openreview.net
Applying reinforcement learning to combinatorial optimization problems is attractive as it
obviates the need for expert knowledge or pre-solved instances. However, it is unrealistic to …