Discriminator Soft Actor Critic without Extrinsic Rewards
This work addresses imitation learning for robotics or simulation tasks, offering an incremental improvement over existing methods like Soft Q imitation learning.
The paper tackles the challenge of imitation learning from limited expert data by proposing Discriminator Soft Actor Critic (DSAC), which replaces constant rewards with an adversarial reward function to improve robustness against distribution shift, achieving efficient learning with only four expert trajectories in PyBullet environments.
It is difficult to be able to imitate well in unknown states from a small amount of expert data and sampling data. Supervised learning methods such as Behavioral Cloning do not require sampling data, but usually suffer from distribution shift. The methods based on reinforcement learning, such as inverse reinforcement learning and generative adversarial imitation learning (GAIL), can learn from only a few expert data. However, they often need to interact with the environment. Soft Q imitation learning addressed the problems, and it was shown that it could learn efficiently by combining Behavioral Cloning and soft Q-learning with constant rewards. In order to make this algorithm more robust to distribution shift, we propose Discriminator Soft Actor Critic (DSAC). It uses a reward function based on adversarial inverse reinforcement learning instead of constant rewards. We evaluated it on PyBullet environments with only four expert trajectories.