SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2408.11052 · ICLR · 2024

ACCELERATING GOAL-CONDITIONED REINFORCE-MENT LEARNING ALGORITHMS AND RESEARCH

Benjamin Eysenbach, Vivek Myers, Łukasz Kuci Ński, Michał Bortkiewicz, Władysław Pałucki, Tadeusz Dziarmaga, Tomasz Arczewski

arXiv · PDF · Open in the Atlas

Code that ran

6 of 9 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (0 of them constructed an object rather than computing a result; 2 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 4 failures of our instrument; 3 no recorded run), measured 2026-11-22; 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/2408.11052.json · badge: /api/ran/2408.11052.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
michalbortkiewicz/jaxgcrl canonical (classifier label)
classifier basis: census
6 ran · 2 no recorded run
MichalBortkiewicz/JaxGCRL canonical (classifier label)
classifier basis: classifier_2025
0 ran · 1 no recorded run
FunctionStatusWhere it lives
apply_logit_scale Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
michalbortkiewicz/jaxgcrl/jaxgcrl/agents/crl/losses.py
code served (permissive licence) · get_code("e848a7e1eecab2c4")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ArrayImpl, ndim 2, dtype family float, shape (2, 4) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import jax.numpy as jnp

MAX_LOGIT_SCALE = 100.0

# Create config dict with energy_fn key
config = {
    'energy_fn': 'cosine'
}

# Create critic_params dict with log_logit_scale
critic_params = {
    'log_logit_scale': jnp.array(0.5)
}

# Create energy array (small tensor)
energy = jnp.array([[1.0, 0.5, -0.3, 0.8],
                     [0.2, -0.5, 0.9, 0.1]])

ARGS = {
    'config': config,
    'critic_params': critic_params,
    'energy': energy
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, ndim 2, dtype family float, shape (2, 4) — shape (2, 4) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
contrastive_loss_fn Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
michalbortkiewicz/jaxgcrl/jaxgcrl/agents/crl/losses.py
code served (permissive licence) · get_code("ba21bcd0d0467788")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ArrayImpl, ndim 0, dtype family float, shape () as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import jax
import jax.numpy as jnp
import numpy as np

# Create a small logits tensor for testing
# The function uses logits in matrix operations (diag, logsumexp on axes)
# So we need a 2D tensor (square or near-square for meaningful diag extraction)
batch_size = 2
logits = jnp.ones((batch_size, batch_size)) * 0.5

# Create ARGS dict with all parameters
ARGS = {
    'name': 'fwd_infonce',
    'logits': logits
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, ndim 0, dtype family float, shape () — shape () as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
energy_fn Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
michalbortkiewicz/jaxgcrl/jaxgcrl/agents/crl/losses.py
code served (permissive licence) · get_code("d2c4953957f1533f")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ArrayImpl, ndim 1, dtype family float, shape (2,) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import jax.numpy as jnp

# Create small test tensors with distinct dimensions
# Batch size: 2, Feature dimension: 4
x = jnp.array([[1.0, 2.0, 3.0, 4.0],
                [5.0, 6.0, 7.0, 8.0]])

y = jnp.array([[0.5, 1.5, 2.5, 3.5],
                [4.5, 5.5, 6.5, 7.5]])

# Test with 'norm' as the default energy function name
name = 'norm'

ARGS = {
    'name': name,
    'x': x,
    'y': y
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, ndim 1, dtype family float, shape (2,) — shape (2,) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
find Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
michalbortkiewicz/jaxgcrl/jaxgcrl/envs/ant_ball_maze.py
code served (permissive licence) · get_code("55de8ec7d2fb9e0e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ArrayImpl, ndim 2, dtype family int, shape (5, 2) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import jax.numpy as jnp
import numpy as np

# Construct arguments for the find function
# The function searches for a specific object in a 2D structure and returns scaled coordinates

# Create a 2D structure (list of lists) containing some objects
structure = [
    [1, 2, 3, 1],
    [4, 1, 5, 2],
    [1, 6, 1, 7]
]

# Size scaling factor - used to scale the indices
size_scaling = 10

# Object to find in the structure
obj = 1

ARGS = {
    "structure": structure,
    "size_scaling": size_scaling,
    "obj": obj
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, ndim 2, dtype family int, shape (5, 2) — shape (5, 2) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
find_goals Ran
ran, honoured the contract we drafted
michalbortkiewicz/jaxgcrl/jaxgcrl/envs/ant_maze.py
code served (permissive licence) · get_code("12acaee1021d466e")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ArrayImpl, ndim 2, dtype family int, shape (4, 2) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import jax.numpy as jnp

# Define the structure as a 2D list/array with some 'g' characters for goals
structure = [
    ['g', '.', '.', '.'],
    ['.', '.', 'g', '.'],
    ['.', '.', '.', '.'],
    ['g', '.', '.', 'g']
]

# Define size_scaling as a positive integer
size_scaling = 2

ARGS = {
    'structure': structure,
    'size_scaling': size_scaling
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, ndim 2, dtype family int, shape (4, 2) — shape (4, 2) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
find_starts Ran
ran, honoured the contract we drafted
michalbortkiewicz/jaxgcrl/jaxgcrl/envs/ant_maze.py
code served (permissive licence) · get_code("8acce921c5f5d9f6")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API) and returned ArrayImpl, ndim 2, dtype family int, shape (3, 2) as a later run recorded it. The image it ran in was not recorded. If your copy returns the same type, ndim and shape on these inputs, it matched what we observed; nothing here checks that. It does NOT mean the output is correct: nothing has compared it to the paper.

import jax.numpy as jnp

RESET = R = 'r'

# Create a small 2D structure grid containing at least one 'r' character
# The function iterates through structure[i][j], so we need a 2D list/array
structure = [
    ['r', '.', '.'],
    ['.', 'r', '.'],
    ['.', '.', 'r']
]

# size_scaling is used as a multiplier for indices
size_scaling = 2

ARGS = {
    'structure': structure,
    'size_scaling': size_scaling
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, ndim 2, dtype family int, shape (3, 2) — shape (3, 2) as a later run recorded it
dtype family
int means an integer dtype; float means any other dtype the instrument saw, bool and complex included (fixture_check.py records only these two families)
Environment
image not recorded; re-verify against syntology-harvest-sandbox:v5-deps; not recorded for this sample: the harness installs a cuda-to-cpu shim (cuda_shim.py) before every sample and no run records whether a sample needed it; seed: none by the harness; any seed is in the setup block
flatten_batch Not yet run
unverified (harvested, no recorded run)
MichalBortkiewicz/JaxGCRL/jaxgcrl/agents/crl/crl.py
code served (permissive licence) · get_code("e9ec954a0a71c3f3")
How it ran

no invocation recorded.

make_maze Not yet run
unverified (harvested, no recorded run)
michalbortkiewicz/jaxgcrl/jaxgcrl/envs/ant_ball_maze.py
code served (permissive licence) · get_code("b9fed04b3ce8f205")
How it ran

no invocation recorded.

make_maze Not yet run
unverified (harvested, no recorded run)
michalbortkiewicz/jaxgcrl/jaxgcrl/envs/ant_maze.py
code served (permissive licence) · get_code("c4df795fab23a2d9")
How it ran

no invocation recorded.

Also present in other papers' code

Of 9 samples lifted from this paper's linked repositories, 9 carry a function identity and 0 do not. Of the 9 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 9 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.

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 2026-11-22.

Repositories linked to this paper

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

Abstract

Self-supervision has the potential to transform reinforcement learning (RL), paralleling the breakthroughs it has enabled in other areas of machine learning. While self-supervised learning in other domains aims to find patterns in a fixed dataset, self-supervised goal-conditioneds reinforcement learning (GCRL) agents discover new behaviors by learning from the goals achieved during unstructured interaction with the environment. However, these methods have failed to see similar success, both due to a lack of data from slow environment simulations as well as a lack of stable algorithms. We take a step toward addressing both of these issues by releasing a high-performance codebase and benchmark (JaxGCRL) for self-supervised GCRL, enabling researchers to train agents for millions of environment steps in minutes on a single GPU. By utilizing GPU-accelerated replay buffers, environments, and a stable contrastive RL algorithm, we reduce training time by up to 22×. Additionally, we assess key design choices in contrastive RL, identifying those that most effectively stabilize and enhance training performance. With this approach, we provide a foundation for future research in self-supervised GCRL, enabling researchers to quickly iterate on new ideas and evaluate them in diverse and challenging environments. Code: https://github.com/MichalBortkiewicz/JaxGCRL.

For agents

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

get_harvested_code_for_paper("2408.11052")
get_code_for_paper("2408.11052")
have("2408.11052")

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

curl https://syntology.ai/api/ran/2408.11052.json

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

[![Syntology run record](https://syntology.ai/api/ran/2408.11052.svg)](https://syntology.ai/paper/2408.11052)

Connect an agent — have() is free.