Policy Manifold Search: Exploring the Manifold Hypothesis for Diversity-based Neuroevolution
This work addresses the problem of scaling neuroevolution for policy search in robotics and AI, offering a novel approach that could enhance efficiency and diversity, though it appears incremental within the quality-diversity framework.
The paper tackles the scalability issue of neuroevolution in high-dimensional parameter spaces by hypothesizing a low-dimensional manifold of diverse, useful policies and proposes a method that searches in a learned representation space, achieving improved performance on continuous-control tasks compared to diversity-based baselines.
Neuroevolution is an alternative to gradient-based optimisation that has the potential to avoid local minima and allows parallelisation. The main limiting factor is that usually it does not scale well with parameter space dimensionality. Inspired by recent work examining neural network intrinsic dimension and loss landscapes, we hypothesise that there exists a low-dimensional manifold, embedded in the policy network parameter space, around which a high-density of diverse and useful policies are located. This paper proposes a novel method for diversity-based policy search via Neuroevolution, that leverages learned representations of the policy network parameters, by performing policy search in this learned representation space. Our method relies on the Quality-Diversity (QD) framework which provides a principled approach to policy search, and maintains a collection of diverse policies, used as a dataset for learning policy representations. Further, we use the Jacobian of the inverse-mapping function to guide the search in the representation space. This ensures that the generated samples remain in the high-density regions, after mapping back to the original space. Finally, we evaluate our contributions on four continuous-control tasks in simulated environments, and compare to diversity-based baselines.