Adversarial Motion Priors Make Good Substitutes for Complex Reward Functions
This addresses the tedious and non-generalizable process of designing reward functions for reinforcement learning in robotics, offering a more efficient alternative for practitioners.
The paper tackles the problem of training simulated agents with under-specified reward functions, which often leads to unnatural behaviors, by proposing to replace complex hand-designed rewards with learned 'style rewards' from motion capture data. The result is that this approach produces naturalistic and energy-efficient locomotion strategies on a real quadrupedal robot, with effective style rewards learned from just a few seconds of data.
Training a high-dimensional simulated agent with an under-specified reward function often leads the agent to learn physically infeasible strategies that are ineffective when deployed in the real world. To mitigate these unnatural behaviors, reinforcement learning practitioners often utilize complex reward functions that encourage physically plausible behaviors. However, a tedious labor-intensive tuning process is often required to create hand-designed rewards which might not easily generalize across platforms and tasks. We propose substituting complex reward functions with "style rewards" learned from a dataset of motion capture demonstrations. A learned style reward can be combined with an arbitrary task reward to train policies that perform tasks using naturalistic strategies. These natural strategies can also facilitate transfer to the real world. We build upon Adversarial Motion Priors -- an approach from the computer graphics domain that encodes a style reward from a dataset of reference motions -- to demonstrate that an adversarial approach to training policies can produce behaviors that transfer to a real quadrupedal robot without requiring complex reward functions. We also demonstrate that an effective style reward can be learned from a few seconds of motion capture data gathered from a German Shepherd and leads to energy-efficient locomotion strategies with natural gait transitions.