Tag
1 article
Google DeepMind researchers have used semantic evolution to create non-intuitive variants of VAD-CFR and SHOR-PSRO algorithms, significantly improving algorithmic convergence in multi-agent reinforcement learning.