Proximal Policy Optimization Algorithms

Benchmark Model Rank Results
neural-architecture-search-on-nats-benchPPO (Schulman et al., 2017)#1Test Accuracy: 44.95
neural-architecture-search-on-nats-bench-1PPO (Schulman et al., 2017)#2Test Accuracy: 94.02
neural-architecture-search-on-nats-bench-2PPO (Schulman et al., 2017)#1Test Accuracy: 71.68
openai-gym-on-ant-v4PPO#5Average Return: 608.97
openai-gym-on-halfcheetah-v4PPO#5Average Return: 6006.11
openai-gym-on-hopper-v4PPO#5Average Return: 790.77
openai-gym-on-humanoid-v4PPO#3Average Return: 925.89
openai-gym-on-walker2d-v4PPO#4Average Return: 2739.81