Policy Optimization by Genetic Distillation

Tanmay Gangwani, Jian Peng

Feb 15, 2018 (modified: Feb 24, 2018) ICLR 2018 Conference Blind Submission readers: everyone Show Bibtex
  • Abstract: Genetic algorithms have been widely used in many practical optimization problems. Inspired by natural selection, operators, including mutation, crossover and selection, provide effective heuristics for search and black-box optimization. However, they have not been shown useful for deep reinforcement learning, possibly due to the catastrophic consequence of parameter crossovers of neural networks. Here, we present Genetic Policy Optimization (GPO), a new genetic algorithm for sample-efficient deep policy optimization. GPO uses imitation learning for policy crossover in the state space and applies policy gradient methods for mutation. Our experiments on MuJoCo tasks show that GPO as a genetic algorithm is able to provide superior performance over the state-of-the-art policy gradient methods and achieves comparable or higher sample efficiency.
  • TL;DR: Genetic algorithms based approach for optimizing deep neural network policies
  • Keywords: Genetic algorithms, deep reinforcement learning, imitation learning

Loading