Diversity is All You Need: Learning Skills without a Reward Function
This addresses the challenge of exploration and data efficiency in reinforcement learning for robotics and AI systems, offering a novel unsupervised pretraining approach.
The paper tackles the problem of learning useful skills without supervision in reinforcement learning, proposing DIAYN, a method that maximizes an information theoretic objective to learn diverse skills like walking and jumping in simulated robotic tasks, and shows it can solve benchmark tasks without true rewards and improve exploration and data efficiency.
Intelligent creatures can explore their environments and learn useful skills without supervision. In this paper, we propose DIAYN ('Diversity is All You Need'), a method for learning useful skills without a reward function. Our proposed method learns skills by maximizing an information theoretic objective using a maximum entropy policy. On a variety of simulated robotic tasks, we show that this simple objective results in the unsupervised emergence of diverse skills, such as walking and jumping. In a number of reinforcement learning benchmark environments, our method is able to learn a skill that solves the benchmark task despite never receiving the true task reward. We show how pretrained skills can provide a good parameter initialization for downstream tasks, and can be composed hierarchically to solve complex, sparse reward tasks. Our results suggest that unsupervised discovery of skills can serve as an effective pretraining mechanism for overcoming challenges of exploration and data efficiency in reinforcement learning.