SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2310.02782 · NeurIPS · 2023

Discovering General Reinforcement Learning Algorithms with Adversarial Environment Design

Shimon Whiteson, Jack Parker-Holder, Jakob Foerster, Gregory Farquhar, Risto Vuorio, Minqi Jiang, Chris Lu, Matthew Jackson

arXiv · PDF · Open in the Atlas

Code that ran

1 of 1 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; 0 honoured the contract we drafted, 0 violated it and 0 have no contract verdict; 1 failure of our instrument; 0 no recorded run), measured 2026-11-22; 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/2310.02782.json · badge: /api/ran/2310.02782.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
EmptyJackson/groove canonical (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
parse_args Ran
ran; our contract draft was wrong, not the code
EmptyJackson/groove/experiments/parse_args.py
code served (permissive licence) · get_code("00007ddfbae017fa")
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) and returned Namespace. The image it ran in was not recorded. If your copy returns the same type 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 sys
import argparse

# Define ARGS dict with all parameters for parse_args
ARGS = {
    'cmd_args': [
        '--debug',
        '--debug_nans',
        '--env_name', 'GridWorld-v0',
        '--env_mode', 'all_shortlife',
        '--env_workers', '64',
        '--seed', '0',
        '--train_steps', '30000',
        '--num_agents', '512',
        '--num_mini_batches', '16',
        '--log',
        '--wandb_project', 'test_project',
        '--wandb_entity', 'test_entity',
        '--wandb_group', 'debug',
        '--train_rollout_len', '20',
        '--gamma', '0.99',
        '--gae_lambda', '0.95',
        '--entropy_coeff', '0.01',
        '--lpg_embedding_net_width', '16',
        '--lpg_gru_width', '256',
        '--lpg_target_width', '8',
        '--lpg_agent_target_coeff', '0.5',
        '--lpg_opt', 'Adam',
        '--lpg_learning_rate', '0.0001',
        '--num_agent_updates', '5',
        '--lpg_max_grad_norm', '0.5',
        '--lpg_policy_entropy_coeff', '0.05',
        '--lpg_target_entropy_coeff', '0.001',
        '--lpg_policy_l2_coeff', '0.005',
        '--lpg_target_l2_coeff', '0.001',
        '--use_es',
        '--es_lrate_decay', '0.999',
        '--es_lrate_limit', '0.00001',
        '--es_sigma_init', '0.1',
        '--es_sigma_decay', '1.0',
        '--es_sigma_limit', '0.1',
        '--es_mean_decay', '0.0',
        '--lifetime_conditioning',
        '--buffer_size', '4000',
        '--score_function', 'random',
        '--p_replay', '0.5',
        '--score_transform', 'rank',
        '--score_temperature', '1.0',
    ]
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
Imports
none outside the standard library — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Namespace — a Namespace output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a Namespace output the recorded ndim 0 and dtype family float describe that coercion, not the output, and are not stated
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

Also present in other papers' code

Of 1 samples lifted from this paper's linked repositories, 1 carry a function identity and 0 do not. Of the 1 distinct functions with an identity, 1 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 1 later) and 0 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
parse_args 1 other paper
1 later
2402.05828 (2024-02)
Discovering Temporally-Aware Reinforcement Learning Algorithms

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

The past decade has seen vast progress in deep reinforcement learning (RL) on the back of algorithms manually designed by human researchers. Recently, it has been shown that it is possible to meta-learn update rules, with the hope of discovering algorithms that can perform well on a wide range of RL tasks. Despite impressive initial results from algorithms such as Learned Policy Gradient (LPG), there remains a generalization gap when these algorithms are applied to unseen environments. In this work, we examine how characteristics of the meta-training distribution impact the generalization performance of these algorithms. Motivated by this analysis and building on ideas from Unsupervised Environment Design (UED), we propose a novel approach for automatically generating curricula to maximize the regret of a meta-learned optimizer, in addition to a novel approximation of regret, which we name algorithmic regret (AR). The result is our method, General RL Optimizers Obtained Via Environment Design (GROOVE). In a series of experiments, we show that GROOVE achieves superior generalization to LPG, and evaluate AR against baseline metrics from UED, identifying it as a critical component of environment design in this setting. We believe this approach is a step towards the discovery of truly general RL algorithms, capable of solving a wide range of real-world environments.

For agents

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

get_harvested_code_for_paper("2310.02782")
get_code_for_paper("2310.02782")
have("2310.02782")

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

curl https://syntology.ai/api/ran/2310.02782.json

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

[![Syntology run record](https://syntology.ai/api/ran/2310.02782.svg)](https://syntology.ai/paper/2310.02782)

Connect an agent — have() is free.