SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2007.04309 · ICLR · 2021

Self-Supervised Policy Adaptation during Deployment

Pieter Abbeel, Lerrel Pinto, Alexei Efros, Xiaolong Wang, Nicklas Hansen, Rishabh Jangir, Yu Sun, Guillem Alenyà, Uc Diego

arXiv · PDF · Open in the Atlas

Code that ran

16 of 21 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (9 of them constructed an object rather than computing a result; 0 honoured the contract we drafted, 0 violated it and 10 have no contract verdict; 6 failures of our instrument; 5 no recorded run), measured 2026-11-21; repository links from the Papers with Code archive attributed under CC BY-SA, no affiliation; bodies inline only under a permissive licence as GitHub detects it. Not a reproduction of the paper's results.

"Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results. This record, dated, as JSON: /api/ran/2007.04309.json · badge: /api/ran/2007.04309.svg

Each repository's role is a label: a classifier's output recorded on the link, or, where no role is recorded, what the Papers with Code archive marked. The line under it names its basis. Neither says who wrote the code.

RepositoryRoleRan
nicklashansen/policy-adaptation-during-deployment community
Papers with Code archive
12 ran · 4 no recorded run
joshnroy/contrastive-pad community
Papers with Code archive
2 ran · 1 no recorded run
copy not recorded community
not in the paper's declared repositories
2 ran · 0 no recorded run
FunctionStatusWhere it lives
Actor Ran
ran on a synthesized input · constructed an object; no result was computed
nicklashansen/policy-adaptation-during-deployment/src/agent/agent.py
pointer only (licence: NONE) · get_code("d526a377580646ff")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, nicklashansen/policy-adaptation-during-deployment when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

BarlowFunction Ran
ran on a synthesized input · constructed an object; no result was computed
joshnroy/contrastive-pad/src/agent/agent.py
pointer only (licence: NONE) · get_code("554f154e05c8c199")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, joshnroy/contrastive-pad when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

CURL Ran
ran on a synthesized input · constructed an object; no result was computed
nicklashansen/policy-adaptation-during-deployment/src/agent/agent.py
pointer only (licence: NONE) · get_code("75b91cb1cba4f86b")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, nicklashansen/policy-adaptation-during-deployment when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

CenterCrop Ran
ran on a synthesized input · constructed an object; no result was computed
nicklashansen/policy-adaptation-during-deployment/src/agent/agent.py
pointer only (licence: NONE) · get_code("16eb246116ba494b")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, nicklashansen/policy-adaptation-during-deployment when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

Critic Ran
ran on a synthesized input · constructed an object; no result was computed
nicklashansen/policy-adaptation-during-deployment/src/agent/agent.py
pointer only (licence: NONE) · get_code("e70accfa4161d131")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, nicklashansen/policy-adaptation-during-deployment when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

InvFunction Ran
ran on a synthesized input · constructed an object; no result was computed
nicklashansen/policy-adaptation-during-deployment/src/agent/agent.py
pointer only (licence: NONE) · get_code("fb1526043e3ddf0f")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, nicklashansen/policy-adaptation-during-deployment when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

NormalizeImg Ran
ran on a synthesized input
nicklashansen/policy-adaptation-during-deployment/src/agent/agent.py
pointer only (licence: NONE) · get_code("21c9f18b03dc202c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, nicklashansen/policy-adaptation-during-deployment when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

PixelEncoder Ran
ran on a synthesized input · constructed an object; no result was computed
nicklashansen/policy-adaptation-during-deployment/src/agent/agent.py
pointer only (licence: NONE) · get_code("77cffce9f453998c")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, nicklashansen/policy-adaptation-during-deployment when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

QFunction Ran
ran on a synthesized input · constructed an object; no result was computed
nicklashansen/policy-adaptation-during-deployment/src/agent/agent.py
pointer only (licence: NONE) · get_code("5447123791608219")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, nicklashansen/policy-adaptation-during-deployment when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

RotFunction Ran
ran on a synthesized input · constructed an object; no result was computed
nicklashansen/policy-adaptation-during-deployment/src/agent/agent.py
pointer only (licence: NONE) · get_code("1e6ff078220508ca")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, nicklashansen/policy-adaptation-during-deployment when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

_rotate_single_with_label Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
nicklashansen/policy-adaptation-during-deployment/src/agent/agent.py
pointer only (licence: NONE) · get_code("1a680b614b0ffdb5")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, nicklashansen/policy-adaptation-during-deployment when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

aug Ran
ran; our contract draft was wrong, not the code
joshnroy/contrastive-pad/src/agent/agent.py
pointer only (licence: NONE) · get_code("e432c8c8eddcb3b7")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, joshnroy/contrastive-pad when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

gaussian_logprob Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from bu-depend-lab/dribo
pointer only · get_code("daa7b3a0355eb7c3")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

make_encoder Ran
ran; our contract draft was wrong, not the code
nicklashansen/policy-adaptation-during-deployment/src/agent/agent.py
pointer only (licence: NONE) · get_code("6dcb9b4b8ec4aabd")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, nicklashansen/policy-adaptation-during-deployment when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

rotate Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
nicklashansen/policy-adaptation-during-deployment/src/agent/agent.py
pointer only (licence: NONE) · get_code("638f9c8f01f54ad1")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, nicklashansen/policy-adaptation-during-deployment when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

squash Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from bu-depend-lab/dribo
pointer only · get_code("e4f442d1770eacd1")
How it ran

The body of this sample is withheld: this paper's copy of the code was not recorded (its link names no repository or source path), so no licence was read for it. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. A run is recorded.

SacSSAgent Not yet run
unverified (harvested, no recorded run)
nicklashansen/policy-adaptation-during-deployment/src/agent/agent.py
pointer only (licence: NONE) · get_code("6563fe457e0a49ef")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, nicklashansen/policy-adaptation-during-deployment when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

SacSSAgent Not yet run
unverified (harvested, no recorded run)
joshnroy/contrastive-pad/src/agent/agent.py
pointer only (licence: NONE) · get_code("b79c90b2fd9d2a85")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, joshnroy/contrastive-pad when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

soft_update_params Not yet run
unverified (harvested, no recorded run)
nicklashansen/policy-adaptation-during-deployment/src/agent/agent.py
pointer only (licence: NONE) · get_code("bc0768cc0f4f49c7")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, nicklashansen/policy-adaptation-during-deployment when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

tie_weights Not yet run
unverified (harvested, no recorded run)
nicklashansen/policy-adaptation-during-deployment/src/agent/agent.py
pointer only (licence: NONE) · get_code("fc88ef6a629962e5")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, nicklashansen/policy-adaptation-during-deployment when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

weight_init Not yet run
unverified (harvested, no recorded run)
nicklashansen/policy-adaptation-during-deployment/src/agent/agent.py
pointer only (licence: NONE) · get_code("48b3228cfe8d98ae")
How it ran

The body of this sample is withheld: GitHub detected no licence in this paper's repository, nicklashansen/policy-adaptation-during-deployment when it was checked. The setup block that called it is withheld with it, because it is written against that body. source_path and origin say where the code lives, to read under the repository's own terms. No invocation recorded.

Also present in other papers' code

Of 21 samples lifted from this paper's linked repositories, 21 carry a function identity and 0 do not. Of the 21 distinct functions with an identity, 7 are also present in other papers whose code we hold (6 with at least one paper dated earlier, 0 the same month, 7 later) and 14 are only here among the 52,977 papers whose code Syntology holds (as of 2026-09-26).

Only here means: among the 52,977 papers whose code Syntology holds (51,887 with lifted samples, 1,090 through their assemblies only), as of 2026-09-26, no other paper's lifted samples or current assembly bricks contain an identical function body after normalisation.

FunctionAlso present inEarliest-datedOthers named
gaussian_logprob
brick of this paper's assembly: helper
15 other papers
3 dated earlier, 12 later · 3 only as an assembly brick
1910.01741 (2019-10)
Improving Sample Efficiency in Model-Free Reinforcement Learning from Images
2004.04136 (2020-04)
CURL: Contrastive Unsupervised Representations for Reinforcement Learning
2004.14990 (2020-04)
Reinforcement Learning with Augmented Data
and 12 more not named here
soft_update_params
brick of this paper's assembly: helper
15 other papers
2 dated earlier, 13 later · 15 only as an assembly brick
2004.13649 (2020-04)
Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixels
2004.14990 (2020-04)
Reinforcement Learning with Augmented Data
2103.04551 (2021-03)
Behavior From the Void: Unsupervised Active Pre-Training
and 12 more not named here
squash
brick of this paper's assembly: helper
14 other papers
3 dated earlier, 11 later · 3 only as an assembly brick
1910.01741 (2019-10)
Improving Sample Efficiency in Model-Free Reinforcement Learning from Images
2004.04136 (2020-04)
CURL: Contrastive Unsupervised Representations for Reinforcement Learning
2004.14990 (2020-04)
Reinforcement Learning with Augmented Data
and 11 more not named here
CURL
brick of this paper's assembly: loss, model
3 other papers
2 dated earlier, 1 later · 2 only as an assembly brick
2004.04136 (2020-04)
CURL: Contrastive Unsupervised Representations for Reinforcement Learning
2004.14990 (2020-04)
Reinforcement Learning with Augmented Data
2205.00943 (2022-05)
CCLF: A Contrastive-Curiosity-Driven Learning Framework for Sample-Efficient Reinforcement Learning
QFunction
brick of this paper's assembly: model
3 other papers
1 dated earlier, 2 later · 3 only as an assembly brick
2004.14990 (2020-04)
Reinforcement Learning with Augmented Data
2205.00943 (2022-05)
CCLF: A Contrastive-Curiosity-Driven Learning Framework for Sample-Efficient Reinforcement Learning
2505.16734 (2025-05)
Maximum Total Correlation Reinforcement Learning
tie_weights
brick of this paper's assembly: helper
3 other papers
2 dated earlier, 1 later · 3 only as an assembly brick
2004.13649 (2020-04)
Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixels
2004.14990 (2020-04)
Reinforcement Learning with Augmented Data
2205.00943 (2022-05)
CCLF: A Contrastive-Curiosity-Driven Learning Framework for Sample-Efficient Reinforcement Learning
weight_init
brick of this paper's assembly: helper
1 other paper
1 later · 1 only as an assembly brick
2305.14133 (2023-05)
Conditional Mutual Information for Disentangled Representations in Reinforcement Learning

Presence, not direction: an identical function body in two papers' repositories does not say who wrote it, which came from which, or that either paper's method rests on it. Matching is exact after normalisation, so a renamed or edited function is not matched. Dates are when papers appeared, not when code was written. Only .py files the harvester lifted are covered; notebooks are not read. A new-style arXiv id YYMM.NNNN dates a paper to the month 20YY-MM; otherwise Paper.year dates it to the year; otherwise it is undated. Dates are when the paper appeared, not when the code was written. Read 2026-11-21.

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

In most real world scenarios, a policy trained by reinforcement learning in one environment needs to be deployed in another, potentially quite different environment. However, generalization across different environments is known to be hard. A natural solution would be to keep training after deployment in the new environment, but this cannot be done if the new environment offers no reward signal. Our work explores the use of self-supervision to allow the policy to continue training after deployment without using any rewards. While previous methods explicitly anticipate changes in the new environment, we assume no prior knowledge of those changes yet still obtain significant improvements. Empirical evaluations are performed on diverse simulation environments from DeepMind Control suite and ViZDoom, as well as real robotic manipulation tasks in continuously changing environments, taking observations from an uncalibrated camera. Our method improves generalization in 31 out of 36 environments across various tasks and outperforms domain randomization on a majority of environments. 1

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2007.04309")
get_code_for_paper("2007.04309")
have("2007.04309")

The run record, dated, one paper per request, free:

curl https://syntology.ai/api/ran/2007.04309.json

A badge for a README (the split and the date, never a ratio):

[![Syntology run record](https://syntology.ai/api/ran/2007.04309.svg)](https://syntology.ai/paper/2007.04309)

Connect an agent — have() is free.