Open-L2O: A Comprehensive and Reproducible Benchmark for Learning to Optimize Algorithms

L2O/NCO codes from CIAM Group at SUSTech, Shenzhen, China

MetaBox: A Benchmark Platform for Meta-Black-Box Optimization with Reinforcement Learning (https://arxiv.org/abs/2310.08252)

[ICML 2024] "MVMoE: Multi-Task Vehicle Routing Solver with Mixture-of-Experts"

[ICML 2023] "Towards Omni-generalizable Neural Methods for Vehicle Routing Problems"

This repo implements our paper, "Learning to Search Feasible and Infeasible Regions of Routing Problems with Flexible Neural k-Opt", which has been accepted at NeurIPS 2023.

A collection of MetaBBO papers and code sources

[NeurIPS 2020 Spotlight Oral] "Training Stronger Baselines for Learning to Optimize", Tianlong Chen*, Weiyi Zhang*, Jingyang Zhou, Shiyu Chang, Sijia Liu, Lisa Amini, Zhangyang Wang

Optim4RL is a Jax framework of learning to optimize for reinforcement learning.

[ICLR 2022] "Bayesian Modeling and Uncertainty Quantification for Learning to Optimize: What, Why, and How" by Yuning You, Yue Cao, Tianlong Chen, Zhangyang Wang, Yang Shen

[ICML 2023] "Learning to Optimize Differentiable Games" by Xuxi Chen, Nelson Vadori, Tianlong Chen, Zhangyang Wang

[ICLR 2023] "M-L2O: Towards Generalizable Learning-to-Optimize by Test-Time Fast Self-Adaptation" by Junjie Yang, Xuxi Chen, Tianlong Chen, Zhangyang Wang, Yingbin Liang

[ECCV 2022] "Scalable Learning to Optimize: A Learned Optimizer Can Train Big Models" by Xuxi Chen, Tianlong Chen, Yu Cheng, Weizhu Chen, Ahmed Awadallah, and Zhangyang Wang

[ICLR 2022] "Optimizer Amalgamation" by Tianshu Huang, Tianlong Chen, Sijia Liu, Shiyu Chang, Lisa Amini, Zhangyang Wang

Operator splitting can be used to design easy-to-train models for predict-and-optimize tasks, which scale effortlessly to problems with thousands of variables.

TF2 implementation of Learning to learn by gradient descent by gradient descent