Agents with a self-attention “bottleneck” not only can solve these tasks from pixel inputs with only 4000 parameters, but they are also better at generalization.
Redirecting to attentionagent.github.io, where the article resides.
Neuroevolution of Self-Interpretable Agents
Agents with a self-attention “bottleneck” not only can solve these tasks from pixel inputs with only 4000 parameters, but they are also better at generalization.