TASK	DATASET	MODEL	METRIC NAME	METRIC VALUE	GLOBAL RANK
Continuous Control	2D Walker	TRPO	Score	1353.8	# 1
Continuous Control	Acrobot	TRPO	Score	-326	# 1
Continuous Control	Acrobot (limited sensors)	TRPO	Score	-83.3	# 1
Continuous Control	Acrobot (noisy observations)	TRPO	Score	-149.6	# 1
Continuous Control	Acrobot (system identifications)	TRPO	Score	-170.9	# 1
Continuous Control	Ant	TRPO	Score	730.2	# 1
Continuous Control	Ant + Gathering	TRPO	Score	-0.4	# 1
Continuous Control	Ant + Maze	TRPO	Score	0	# 1
Continuous Control	Cart-Pole Balancing	TRPO	Score	4869.8	# 1
Continuous Control	Cart-Pole Balancing (limited sensors)	TRPO	Score	960.2	# 1
Continuous Control	Cart-Pole Balancing (noisy observations)	TRPO	Score	606.2	# 1
Continuous Control	Cart-Pole Balancing (system identifications)	TRPO	Score	980.3	# 1
Continuous Control	Double Inverted Pendulum	TRPO	Score	4412.4	# 1
Continuous Control	Full Humanoid	TRPO	Score	287	# 1
Continuous Control	Half-Cheetah	TRPO	Score	1914	# 1
Continuous Control	Hopper	TRPO	Score	1183.3	# 1
Continuous Control	Inverted Pendulum	TRPO	Score	247.2	# 1
Continuous Control	Inverted Pendulum (limited sensors)	TRPO	Score	4.5	# 1
Continuous Control	Inverted Pendulum (noisy observations)	TRPO	Score	10.4	# 1
Continuous Control	Inverted Pendulum (system identifications)	TRPO	Score	14.1	# 1
Continuous Control	Mountain Car	TRPO	Score	-61.7	# 1
Continuous Control	Mountain Car (limited sensors)	TRPO	Score	-64.2	# 1
Continuous Control	Mountain Car (noisy observations)	TRPO	Score	-60.2	# 1
Continuous Control	Mountain Car (system identifications)	TRPO	Score	-61.6	# 1
Continuous Control	Simple Humanoid	TRPO	Score	269.7	# 1
Continuous Control	Swimmer	TRPO	Score	96	# 1
Continuous Control	Swimmer + Gathering	TRPO	Score	0	# 1
Continuous Control	Swimmer + Maze	TRPO	Score	0	# 1

Badge	Markdown
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-2d-walker)](https://paperswithcode.com/sota/continuous-control-on-2d-walker?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-acrobot)](https://paperswithcode.com/sota/continuous-control-on-acrobot?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-acrobot-limited-sensors)](https://paperswithcode.com/sota/continuous-control-on-acrobot-limited-sensors?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-acrobot-noisy)](https://paperswithcode.com/sota/continuous-control-on-acrobot-noisy?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-acrobot-system)](https://paperswithcode.com/sota/continuous-control-on-acrobot-system?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-ant)](https://paperswithcode.com/sota/continuous-control-on-ant?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-ant-gathering)](https://paperswithcode.com/sota/continuous-control-on-ant-gathering?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-ant-maze)](https://paperswithcode.com/sota/continuous-control-on-ant-maze?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-cart-pole-balancing)](https://paperswithcode.com/sota/continuous-control-on-cart-pole-balancing?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-cart-pole-balancing-1)](https://paperswithcode.com/sota/continuous-control-on-cart-pole-balancing-1?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-cart-pole-balancing-2)](https://paperswithcode.com/sota/continuous-control-on-cart-pole-balancing-2?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-cart-pole-balancing-3)](https://paperswithcode.com/sota/continuous-control-on-cart-pole-balancing-3?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-double-inverted)](https://paperswithcode.com/sota/continuous-control-on-double-inverted?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-full-humanoid)](https://paperswithcode.com/sota/continuous-control-on-full-humanoid?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-half-cheetah)](https://paperswithcode.com/sota/continuous-control-on-half-cheetah?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-hopper)](https://paperswithcode.com/sota/continuous-control-on-hopper?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-inverted-pendulum)](https://paperswithcode.com/sota/continuous-control-on-inverted-pendulum?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-inverted-pendulum-1)](https://paperswithcode.com/sota/continuous-control-on-inverted-pendulum-1?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-inverted-pendulum-noisy)](https://paperswithcode.com/sota/continuous-control-on-inverted-pendulum-noisy?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-inverted-pendulum-2)](https://paperswithcode.com/sota/continuous-control-on-inverted-pendulum-2?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-mountain-car)](https://paperswithcode.com/sota/continuous-control-on-mountain-car?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-mountain-car-limited)](https://paperswithcode.com/sota/continuous-control-on-mountain-car-limited?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-mountain-car-noisy)](https://paperswithcode.com/sota/continuous-control-on-mountain-car-noisy?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-mountain-car-system)](https://paperswithcode.com/sota/continuous-control-on-mountain-car-system?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-simple-humanoid)](https://paperswithcode.com/sota/continuous-control-on-simple-humanoid?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-swimmer)](https://paperswithcode.com/sota/continuous-control-on-swimmer?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-swimmer-gathering)](https://paperswithcode.com/sota/continuous-control-on-swimmer-gathering?p=benchmarking-deep-reinforcement-learning-for)`
	`[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/benchmarking-deep-reinforcement-learning-for/continuous-control-on-swimmer-maze)](https://paperswithcode.com/sota/continuous-control-on-swimmer-maze?p=benchmarking-deep-reinforcement-learning-for)`

Benchmarking Deep Reinforcement Learning for Continuous Control

22 Apr 2016 · Yan Duan, Xi Chen, Rein Houthooft, John Schulman, Pieter Abbeel ·

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.

PDF Abstract

Code

Add Remove Mark official

rllab/rllab official

2,805

rll/rllab

2,805

openai/rllab

2,805

rlworkgroup/garage

1,730

cbfinn/maml_rl

582

See all 15 implementations

Tasks

Add Remove

Action Triplet Recognition

Atari Games

Benchmarking

Continuous Control

reinforcement-learning

Reinforcement Learning (RL)

Datasets

Introduced in the Paper:

RLLab Framework

Used in the Paper:

MuJoCo

Results from the Paper

Edit

Ranked #1 on Continuous Control on Inverted Pendulum

Get a GitHub badge

Task	Dataset	Model	Metric Name	Metric Value	Global Rank	Benchmark
Continuous Control	2D Walker	TRPO	Score	1353.8	# 1	Compare
Continuous Control	Acrobot	TRPO	Score	-326	# 1	Compare
Continuous Control	Acrobot (limited sensors)	TRPO	Score	-83.3	# 1	Compare
Continuous Control	Acrobot (noisy observations)	TRPO	Score	-149.6	# 1	Compare
Continuous Control	Acrobot (system identifications)	TRPO	Score	-170.9	# 1	Compare
Continuous Control	Ant	TRPO	Score	730.2	# 1	Compare
Continuous Control	Ant + Gathering	TRPO	Score	-0.4	# 1	Compare
Continuous Control	Ant + Maze	TRPO	Score	0	# 1	Compare
Continuous Control	Cart-Pole Balancing	TRPO	Score	4869.8	# 1	Compare
Continuous Control	Cart-Pole Balancing (limited sensors)	TRPO	Score	960.2	# 1	Compare
Continuous Control	Cart-Pole Balancing (noisy observations)	TRPO	Score	606.2	# 1	Compare
Continuous Control	Cart-Pole Balancing (system identifications)	TRPO	Score	980.3	# 1	Compare
Continuous Control	Double Inverted Pendulum	TRPO	Score	4412.4	# 1	Compare
Continuous Control	Full Humanoid	TRPO	Score	287	# 1	Compare
Continuous Control	Half-Cheetah	TRPO	Score	1914	# 1	Compare
Continuous Control	Hopper	TRPO	Score	1183.3	# 1	Compare
Continuous Control	Inverted Pendulum	TRPO	Score	247.2	# 1	Compare
Continuous Control	Inverted Pendulum (limited sensors)	TRPO	Score	4.5	# 1	Compare
Continuous Control	Inverted Pendulum (noisy observations)	TRPO	Score	10.4	# 1	Compare
Continuous Control	Inverted Pendulum (system identifications)	TRPO	Score	14.1	# 1	Compare
Continuous Control	Mountain Car	TRPO	Score	-61.7	# 1	Compare
Continuous Control	Mountain Car (limited sensors)	TRPO	Score	-64.2	# 1	Compare
Continuous Control	Mountain Car (noisy observations)	TRPO	Score	-60.2	# 1	Compare
Continuous Control	Mountain Car (system identifications)	TRPO	Score	-61.6	# 1	Compare
Continuous Control	Simple Humanoid	TRPO	Score	269.7	# 1	Compare
Continuous Control	Swimmer	TRPO	Score	96	# 1	Compare
Continuous Control	Swimmer + Gathering	TRPO	Score	0	# 1	Compare
Continuous Control	Swimmer + Maze	TRPO	Score	0	# 1	Compare

Methods

Add Remove

No methods listed for this paper. Add relevant methods here

Edit Social Preview

Benchmarking Deep Reinforcement Learning for Continuous Control

Code Edit Add Remove Mark official

Tasks Edit Add Remove

Datasets Edit

Results from the Paper Edit

Methods Edit Add Remove

Code

Add Remove Mark official

Tasks

Add Remove

Datasets

Results from the Paper

Edit

Methods

Add Remove