TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Models/CURL

CURL

Reported on 88 benchmarks across 5 tasks · 1 paper · 38 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Playing Games52 results

  • Atari GamesonAtari 2600 James Bond
    Medium Human-Normalized Score· 2020-04-08
    400.1
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 James Bond
    Medium Human-Normalized Score· 2020-04-08
    400.1
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Boxing
    Score· 2020-04-08
    4.8
    best: 100 (MuZero)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Ms. Pacman
    Score· 2020-04-08
    1492.8
    best: 243401.1 (MuZero)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Freeway
    Score· 2020-04-08
    27.9
    best: 34 (TRPO-hash)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Pong
    Score· 2020-04-08
    2.1
    best: 21 (Duel noop)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Krull
    Score· 2020-04-08
    3833.6
    best: 594540 (GDI-H3)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Breakout
    Score· 2020-04-08
    18.2
    best: 864 (GDI-H3(200M frames))
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Frostbite
    Score· 2020-04-08
    924
    best: 631378.53 (MuZero)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Gopher
    Score· 2020-04-08
    801.4
    best: 488830 (GDI-I3)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Amidar
    Score· 2020-04-08
    232.3
    best: 29660.08 (Agent57)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Crazy Climber
    Score· 2020-04-08
    27805.6
    best: 565909.85 (Agent57)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Demon Attack
    Score· 2020-04-08
    834
    best: 787985 (GDI-H3)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Battle Zone
    Score· 2020-04-08
    11208
    best: 934134.88 (Agent57)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Asterix
    Score· 2020-04-08
    524.3
    best: 999999 (GDI-H3)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Kung-Fu Master
    Score· 2020-04-08
    14280
    best: 1666665 (GDI-H3)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Kangaroo
    Score· 2020-04-08
    345.3
    best: 24034.16 (Agent57)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Assault
    Score· 2020-04-08
    543.7
    best: 143972.03 (MuZero)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Alien
    Score· 2020-04-08
    1148.2
    best: 741812.63 (MuZero)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Seaquest
    Score· 2020-04-08
    408
    best: 1000000 (GDI-H3(200M frames))
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Chopper Command
    Score· 2020-04-08
    1198
    best: 999999 (GDI-H3)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 HERO
    Score· 2020-04-08
    6235.1
    best: 114736.26 (Agent57)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Bank Heist
    Score· 2020-04-08
    193.7
    best: 27219.8 (MuZero (Res2 Adam))
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Private Eye
    Score· 2020-04-08
    105.2
    best: 95756 (Go-Explore)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Q*Bert
    Score· 2020-04-08
    1225.6
    best: 580328.14 (Agent57)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Road Runner
    Score· 2020-04-08
    6786.7
    best: 999999 (GDI-H3)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Atari GamesonAtari 2600 Up and Down
    Score· 2020-04-08
    2735.2
    best: 986440 (GDI-I3)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Boxing
    Score· 2020-04-08
    4.8
    best: 100 (MuZero)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Ms. Pacman
    Score· 2020-04-08
    1492.8
    best: 243401.1 (MuZero)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Freeway
    Score· 2020-04-08
    27.9
    best: 34 (TRPO-hash)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Pong
    Score· 2020-04-08
    2.1
    best: 21 (Duel noop)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Krull
    Score· 2020-04-08
    3833.6
    best: 594540 (GDI-H3)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Breakout
    Score· 2020-04-08
    18.2
    best: 864 (GDI-H3(200M frames))
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Frostbite
    Score· 2020-04-08
    924
    best: 631378.53 (MuZero)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Gopher
    Score· 2020-04-08
    801.4
    best: 488830 (GDI-I3)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Amidar
    Score· 2020-04-08
    232.3
    best: 29660.08 (Agent57)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Crazy Climber
    Score· 2020-04-08
    27805.6
    best: 565909.85 (Agent57)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Demon Attack
    Score· 2020-04-08
    834
    best: 787985 (GDI-H3)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Battle Zone
    Score· 2020-04-08
    11208
    best: 934134.88 (Agent57)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Asterix
    Score· 2020-04-08
    524.3
    best: 999999 (GDI-H3)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Kung-Fu Master
    Score· 2020-04-08
    14280
    best: 1666665 (GDI-H3)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Kangaroo
    Score· 2020-04-08
    345.3
    best: 24034.16 (Agent57)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Assault
    Score· 2020-04-08
    543.7
    best: 143972.03 (MuZero)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Alien
    Score· 2020-04-08
    1148.2
    best: 741812.63 (MuZero)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Seaquest
    Score· 2020-04-08
    408
    best: 1000000 (GDI-H3(200M frames))
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Chopper Command
    Score· 2020-04-08
    1198
    best: 999999 (GDI-H3)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 HERO
    Score· 2020-04-08
    6235.1
    best: 114736.26 (Agent57)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Bank Heist
    Score· 2020-04-08
    193.7
    best: 27219.8 (MuZero (Res2 Adam))
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Private Eye
    Score· 2020-04-08
    105.2
    best: 95756 (Go-Explore)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Q*Bert
    Score· 2020-04-08
    1225.6
    best: 580328.14 (Agent57)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Road Runner
    Score· 2020-04-08
    6786.7
    best: 999999 (GDI-H3)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Video GamesonAtari 2600 Up and Down
    Score· 2020-04-08
    2735.2
    best: 986440 (GDI-I3)
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136

Robots12 results

  • Continuous ControlonWalker, walk (DMControl100k)
    Score· 2020-04-08
    403
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Continuous ControlonCartpole, swingup (DMControl100k)
    Score· 2020-04-08
    582
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Continuous ControlonCheetah, run (DMControl500k)
    Score· 2020-04-08
    518
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Continuous ControlonReacher, easy (DMControl500k)
    Score· 2020-04-08
    929
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Continuous ControlonFinger, spin (DMControl100k)
    Score· 2020-04-08
    767
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Continuous ControlonCheetah, run (DMControl100k)
    Score· 2020-04-08
    299
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Continuous ControlonFinger, spin (DMControl500k)
    Score· 2020-04-08
    926
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Continuous ControlonBall in cup, catch (DMControl500k)
    Score· 2020-04-08
    959
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Continuous ControlonReacher, easy (DMControl100k)
    Score· 2020-04-08
    538
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Continuous ControlonWalker, walk (DMControl500k)
    Score· 2020-04-08
    902
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Continuous ControlonCartpole, swingup (DMControl500k)
    Score· 2020-04-08
    841
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • Continuous ControlonBall in cup, catch (DMControl100k)
    Score· 2020-04-08
    769
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136

Methodology12 results

  • 3DonWalker, walk (DMControl100k)
    Score· 2020-04-08
    403
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3DonCartpole, swingup (DMControl100k)
    Score· 2020-04-08
    582
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3DonCheetah, run (DMControl500k)
    Score· 2020-04-08
    518
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3DonReacher, easy (DMControl500k)
    Score· 2020-04-08
    929
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3DonFinger, spin (DMControl100k)
    Score· 2020-04-08
    767
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3DonCheetah, run (DMControl100k)
    Score· 2020-04-08
    299
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3DonFinger, spin (DMControl500k)
    Score· 2020-04-08
    926
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3DonBall in cup, catch (DMControl500k)
    Score· 2020-04-08
    959
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3DonReacher, easy (DMControl100k)
    Score· 2020-04-08
    538
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3DonWalker, walk (DMControl500k)
    Score· 2020-04-08
    902
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3DonCartpole, swingup (DMControl500k)
    Score· 2020-04-08
    841
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3DonBall in cup, catch (DMControl100k)
    Score· 2020-04-08
    769
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136

Medical12 results

  • 3D Face ModellingonWalker, walk (DMControl100k)
    Score· 2020-04-08
    403
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3D Face ModellingonCartpole, swingup (DMControl100k)
    Score· 2020-04-08
    582
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3D Face ModellingonCheetah, run (DMControl500k)
    Score· 2020-04-08
    518
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3D Face ModellingonReacher, easy (DMControl500k)
    Score· 2020-04-08
    929
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3D Face ModellingonFinger, spin (DMControl100k)
    Score· 2020-04-08
    767
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3D Face ModellingonCheetah, run (DMControl100k)
    Score· 2020-04-08
    299
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3D Face ModellingonFinger, spin (DMControl500k)
    Score· 2020-04-08
    926
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3D Face ModellingonBall in cup, catch (DMControl500k)
    Score· 2020-04-08
    959
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3D Face ModellingonReacher, easy (DMControl100k)
    Score· 2020-04-08
    538
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3D Face ModellingonWalker, walk (DMControl500k)
    Score· 2020-04-08
    902
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3D Face ModellingonCartpole, swingup (DMControl500k)
    Score· 2020-04-08
    841
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136
  • 3D Face ModellingonBall in cup, catch (DMControl100k)
    Score· 2020-04-08
    769
    SOTA
    CURL: Contrastive Unsupervised Representations for Reinforcement LearningarXiv:2004.04136