Papers › First return, then explore

First return, then explore

27 Apr 2020arXiv:2004.12919archive 2025-07-28

Adrien Ecoffet, Joost Huizinga, Joel Lehman, Kenneth O. Stanley, Jeff Clune

The promise of reinforcement learning is to solve complex sequential decision problems autonomously by specifying a high-level reward function only. However, reinforcement learning algorithms struggle when, as is often the case, simple and intuitive rewards provide sparse and deceptive feedback. Avoiding these pitfalls requires thoroughly exploring the environment, but creating algorithms that can do so remains one of the central challenges of the field. We hypothesise that the main impediment to effective exploration originates from algorithms forgetting how to reach previously visited states ("detachment") and from failing to first return to a state before exploring from it ("derailment"). We introduce Go-Explore, a family of algorithms that addresses these two challenges directly through the simple principles of explicitly remembering promising states and first returning to such states before intentionally exploring. Go-Explore solves all heretofore unsolved Atari games and surpasses the state of the art on all hard-exploration games, with orders of magnitude improvements on the grand challenges Montezuma's Revenge and Pitfall. We also demonstrate the practical potential of Go-Explore on a sparse-reward pick-and-place robotics task. Additionally, we show that adding a goal-conditioned policy can further improve Go-Explore's exploration efficiency and enable it to handle stochasticity throughout training. The substantial performance gains from Go-Explore suggest that the simple principles of remembering states, returning to them, and exploring from them are a powerful and general approach to exploration, an insight that may prove critical to the creation of truly intelligent learning agents.

PaperPDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2004.12919")

Code

Syntology Ran 2 of 4 code samples harvested from 1 repository linked to this paper; 2 have no recorded run. Of those that ran: 1 ran · honoured contract; 1 ran · our draft was wrong.

By repository: official repository: 4 samples from 1 repository, 2 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

uber-research/go-explore officialmentioned in papermentioned on GitHubtf report
qgallouedec/lge mentioned on GitHubpytorchMIT report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

4 samples harvested; 2 ran; 1 honoured the contract we drafted; 2 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1ran · honoured contract
1ran · our draft was wrong
2unverified

Licence: 4 of the 4 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from uber-research/go-explore. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

getsize uber-research/go-explore/policy_based/goexplore_py/goexplore.py official repository ran · honoured contract fingerprinted licence not identified · pointer only · 7aca6d4823baaf95 · report
seed_pool_wrapper uber-research/go-explore/robustified/goexplore_py/goexplore.py official repository ran · our draft was wrong licence not identified · pointer only · efd5de4203794fee · report
get_downscale uber-research/go-explore/robustified/goexplore_py/goexplore.py official repository unverified licence not identified · pointer only · c0b41660051f8848 · report
run_f_seeded uber-research/go-explore/robustified/goexplore_py/goexplore.py official repository unverified licence not identified · pointer only · 7f626c87007c22d4 · report

Tasks

Atari GamesMontezuma's RevengeReinforcement LearningReinforcement Learning (RL)reinforcement-learning

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Atari Games Atari 2600 Berzerk Go-Explore Score 197376 #1 of 39 Archive leaderboard report
Atari Games Atari 2600 Bowling Go-Explore Score 260 #2 of 44 Archive leaderboard report
Atari Games Atari 2600 Centipede Go-Explore Score 1422628 #1 of 45 Archive leaderboard report
Atari Games Atari 2600 Freeway Go-Explore Score 34 #5 of 59 Archive leaderboard report
Atari Games Atari 2600 Gravitar Go-Explore Score 7588 #4 of 53 Archive leaderboard report
Atari Games Atari 2600 Montezuma's Revenge Go-Explore Score 43791 #1 of 50 Archive leaderboard report
Atari Games Atari 2600 Pitfall! Go-Explore Score 6954 #3 of 23 Archive leaderboard report
Atari Games Atari 2600 Private Eye Go-Explore Score 95756 #1 of 52 Archive leaderboard report
Atari Games Atari 2600 Skiing Go-Explore Score -3660 #13 of 23 Archive leaderboard report
Atari Games Atari 2600 Solaris Go-Explore Score 19671 #2 of 23 Archive leaderboard report
Atari Games Atari 2600 Venture Go-Explore Score 2281 #2 of 55 Archive leaderboard report
Atari Games Atari games Go-Explore Mean Human Normalized Score 4989.94% #4 of 12 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

Go-Explore

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections