Discovery of independently controllable features through autonomous goal setting
Despite recent breakthroughs in artificial intelligence, machine learning agents remain limited to tasks predefined by human engineers. The autonomous and simultaneous discovery and learning of many-tasks in an open world remains very challenging for reinforcement learning algorithms. In this blog post we explore recent advances in developmental learning to tackle the problem of how an agent can learn to represent, imagine and select its own goals.
How Many Random Seeds ?
Reproducibility in Machine Learning and Deep Reinforcement Learning in particular has become a serious issue in the recent years. In this blog post, we present a statistical guide to perform rigorous comparison of RL algorithms.
Bootstrapping Deep RL with Population-Based Diversity Search
Standard deep RL algorithms using continuous actions suffer from inefficient exploration when facing sparse or deceptive reward problems. Here we propose to decouple exploration and exploitation. An exploration algorithm first optimizes for diversity in the space of behaviors. Then, a state-of-the art deep RL algorithm uses the collected trajectories for bootstrapping.