Parametric-Task MAP-Elites
This addresses the limitation of existing multi-task algorithms that only handle finite task sets, enabling optimization across continuous spaces for applications like robotics, though it appears incremental as an extension of MAP-Elites.
The paper tackled the problem of multi-task optimization for continuous task spaces by introducing Parametric-Task MAP-Elites (PT-ME), a black-box algorithm that solves a new task each iteration and uses a local linear regression variation operator, resulting in outperformance over baselines including PPO on toy and robotic simulation problems.
Optimizing a set of functions simultaneously by leveraging their similarity is called multi-task optimization. Current black-box multi-task algorithms only solve a finite set of tasks, even when the tasks originate from a continuous space. In this paper, we introduce Parametric-Task MAP-Elites (PT-ME), a new black-box algorithm for continuous multi-task optimization problems. This algorithm (1) solves a new task at each iteration, effectively covering the continuous space, and (2) exploits a new variation operator based on local linear regression. The resulting dataset of solutions makes it possible to create a function that maps any task parameter to its optimal solution. We show that PT-ME outperforms all baselines, including the deep reinforcement learning algorithm PPO on two parametric-task toy problems and a robotic problem in simulation.