Task-Agnostic Morphology Evolution
This addresses the issue of task-dependent morphologies in deep reinforcement learning, offering a more efficient and generalizable approach for researchers and practitioners in robotics and AI.
The paper tackles the problem of finding agent morphologies that generalize across tasks without task-specific rewards, proposing Task-Agnostic Morphology Evolution (TAME) which evolves morphologies using random actions and an information-theoretic objective, and demonstrates that TAME matches the multi-task performance of task-supervised algorithms in 2D, 3D, and manipulation environments.
Deep reinforcement learning primarily focuses on learning behavior, usually overlooking the fact that an agent's function is largely determined by form. So, how should one go about finding a morphology fit for solving tasks in a given environment? Current approaches that co-adapt morphology and behavior use a specific task's reward as a signal for morphology optimization. However, this often requires expensive policy optimization and results in task-dependent morphologies that are not built to generalize. In this work, we propose a new approach, Task-Agnostic Morphology Evolution (TAME), to alleviate both of these issues. Without any task or reward specification, TAME evolves morphologies by only applying randomly sampled action primitives on a population of agents. This is accomplished using an information-theoretic objective that efficiently ranks agents by their ability to reach diverse states in the environment and the causality of their actions. Finally, we empirically demonstrate that across 2D, 3D, and manipulation environments TAME can evolve morphologies that match the multi-task performance of those learned with task supervised algorithms. Our code and videos can be found at https://sites.google.com/view/task-agnostic-evolution.