Benchmarking Deep Reinforcement Learning for Continuous Control

Code

• github.com/rllab/rllab
  OfficialIn papertf★ 0
• github.com/rll/rllab
  tf★ 3,050
• github.com/rlworkgroup/garage
  tf★ 2,111
• github.com/russellmendonca/maesn_suite
  tf★ 44
• github.com/jachiam/cpo
  none★ 0
• github.com/sisl/event-driven-rllab
  tf★ 0
• github.com/sisl/gail-driver
  none★ 0
• github.com/cathywu/rllab-multiagent
  tf★ 0
• github.com/wyndwarrior/imitation_from_observation
  tf★ 0
• github.com/Dam930/rllab
  tf★ 0

Abstract

Recently, researchers have made significant progress combining the advances in deep learning for learning feature representations with reinforcement learning. Some notable examples include training agents to play Atari games based on raw pixel data and to acquire advanced manipulation skills using raw sensory inputs. However, it has been difficult to quantify progress in the domain of continuous control due to the lack of a commonly adopted benchmark. In this work, we present a benchmark suite of continuous control tasks, including classic tasks like cart-pole swing-up, tasks with very high state and action dimensionality such as 3D humanoid locomotion, tasks with partial observations, and tasks with hierarchical structure. We report novel findings based on the systematic evaluation of a range of implemented reinforcement learning algorithms. Both the benchmark and reference implementations are released at https://github.com/rllab/rllab in order to facilitate experimental reproducibility and to encourage adoption by other researchers.

Tasks

Benchmark Results

Reproductions