Papers › Learning by Cheating

Learning by Cheating

27 Dec 2019arXiv:1912.12294archive 2025-07-28

Dian Chen, Brady Zhou, Vladlen Koltun, Philipp Krähenbühl

Vision-based urban driving is hard. The autonomous system needs to learn to perceive the world and act in it. We show that this challenging learning problem can be simplified by decomposing it into two stages. We first train an agent that has access to privileged information. This privileged agent cheats by observing the ground-truth layout of the environment and the positions of all traffic participants. In the second stage, the privileged agent acts as a teacher that trains a purely vision-based sensorimotor agent. The resulting sensorimotor agent does not have access to any privileged information and does not cheat. This two-stage training procedure is counter-intuitive at first, but has a number of important advantages that we analyze and empirically demonstrate. We use the presented approach to train a vision-based autonomous driving system that substantially outperforms the state of the art on the CARLA benchmark and the recent NoCrash benchmark. Our approach achieves, for the first time, 100% success rate on all tasks in the original CARLA benchmark, sets a new record on the NoCrash benchmark, and reduces the frequency of infractions by an order of magnitude compared to the prior state of the art. For the video that summarizes this work, see https://youtu.be/u9ZCxxD-UUw

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="1912.12294")

Code

Syntology Ran 0 of 7 code samples harvested from 2 repositories linked to this paper; 7 have no recorded run.

By repository: official repository: 6 samples from 1 repository, 0 ran; community (archive-listed): 1 sample from 1 repository, 0 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

dotchen/LearningByCheating officialmentioned on GitHubpytorchMIT report
SimarKareer/legged_gym mentioned on GitHubpytorchNOASSERTION report
bradyz/2020_CARLA_challenge mentioned on GitHub report
deepsense-ai/carla-birdeye-view mentioned on GitHubMIT report
jostl/masters-thesis mentioned on GitHubpytorchMIT report
piazzesiNiccolo/myLbc mentioned on GitHubpytorchMIT report
scope-lab-vu/anti-carla mentioned on GitHub report
zwc662/SequentialAttack mentioned on GitHubpytorchNOASSERTION 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

7 samples harvested; 0 ran; 0 honoured the contract we drafted; 7 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.

7unverified

Licence: 0 of the 7 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 2 repositories linked to this paper, official or community; each sample names its own and says which. “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.

BaselineBranch dotchen/LearningByCheating/bird_view/models/baseline.py official repository unverified MIT (permissive) · 3ad45fbbefbcc580 · report
crop_birdview dotchen/LearningByCheating/bird_view/models/common.py official repository unverified MIT (permissive) · 1eac6683c6404df6 · report
from_file dotchen/LearningByCheating/benchmark/goal_suite.py official repository unverified MIT (permissive) · 663c3ff922453f02 · report
ls_circle dotchen/LearningByCheating/bird_view/models/controller.py official repository unverified MIT (permissive) · 5da19e3e051b29fd · report
select_branch dotchen/LearningByCheating/bird_view/models/common.py official repository unverified MIT (permissive) · 3bbc730e19952eb1 · report
signed_angle dotchen/LearningByCheating/bird_view/models/common.py official repository unverified MIT (permissive) · 680bf10038ec8494 · report
lateral_shift deepsense-ai/carla-birdeye-view/carla_birdeye_view/lanes.py community (archive-listed) unverified MIT (permissive) · edff33b2be73c06a · report

Tasks

Autonomous Driving

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Autonomous Driving CARLA Leaderboard LBC Driving Score 8.94 #17 of 18 Archive leaderboard report
Autonomous Driving CARLA Leaderboard LBC Infraction penalty 0.73 #17 of 18 Archive leaderboard report
Autonomous Driving CARLA Leaderboard LBC Route Completion 17.54 #17 of 18 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

CARLAEntropy RegularizationPPO

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