Repositories

• MetaBox: The first benchmark platform expressly tailored for developing and evaluating MetaBBO-RL methods, which is accepted at NeurIPS 2023.
• Symbol: The python implementation of our paper SYMBOL, which is accepted as a poster paper at ICLR 2024. This is a novel MetaBBO paragidm against the recent proposed ones, refer to the paper for detail.
• Neur-ELA: The official code for Neural Exploratory Landscape Analysis. It is an automatic extraction of optimization status features through a landscape analyser parameterized by a two-stage attention-based neural network, refer to the paper for detail.
• RL-DAS: The official code for Deep Reinforcement Learning for Dynamic Algorithm Selection: A Proof-of-Principle Study on Differential Evolution. It is a DRL-based dynamic algorithm selection framework, refer to the paper for detail.
• GLEET: The official code for Auto-configuring Exploration-Exploitation Tradeoff in Evolutionary Computation via Deep Reinforcement Learning. It is a DRL-based framework that autonomously configures and adapts the exploration-exploitation tradeoff throughout the EC search process., refer to the paper for detail.
• Awesome-MetaBBO: This is a collection of MetaBBO papers and their corresponding code resources.
• Coming soon