SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1809.04474 · 2018

Multi-task Deep Reinforcement Learning with PopArt

arXiv · PDF · Open in the Atlas

Code that ran

3 of 3 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; 3 failures of our instrument; 0 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/1809.04474.json · badge: /api/ran/1809.04474.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
copy not recorded community
not in the paper's declared repositories
2 ran · 0 no recorded run
aluscher/torchbeastpopart extension (classifier label)
classifier basis: classifier_2025
1 ran · 0 no recorded run
FunctionStatusWhere it lives
compute_baseline_loss Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from aypan17/reward-misspecification
pointer only · get_code("e9123707cba84b2c")
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.

compute_entropy_loss Ran
ran; our contract draft was wrong, not the code
this paper's copy was not recorded; identical code first harvested from aypan17/reward-misspecification
pointer only · get_code("a27a9dc97eefda7f")
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.

compute_policy_gradient_loss Ran
ran; our contract draft was wrong, not the code
aluscher/torchbeastpopart/torchbeast/polybeast.py
code served (permissive licence) · get_code("fd36e71a3f45b446")
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 Tensor, 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 torch
import torch.nn.functional as F

torch.manual_seed(0)

# Analyze the function to determine required shapes:
# - logits: flattened to (batch*time, num_actions), so needs shape (batch, time, num_actions)
# - actions: flattened to (batch*time*something), then viewed back to actions shape
# - advantages: must match actions shape after viewing

# Let's use:
# batch_size = 2
# time_steps = 3
# num_actions = 4

batch_size = 2
time_steps = 3
num_actions = 4

# logits shape: (batch, time, num_actions)
logits = torch.randn(batch_size, time_steps, num_actions)

# actions shape: (batch, time, 1) - this will be flattened to (batch*time*1,)
# and needs to contain valid action indices (0 to num_actions-1)
actions = torch.randint(0, num_actions, (batch_size, time_steps, 1))

# advantages shape: must match actions shape (batch, time, 1)
advantages = torch.randn(batch_size, time_steps, 1)

ARGS = {
    "logits": logits,
    "actions": actions,
    "advantages": advantages,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type Tensor, 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

Also present in other papers' code

Of 3 samples lifted from this paper's linked repositories, 3 carry a function identity and 0 do not. Of the 3 distinct functions with an identity, 2 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 2 later) and 1 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
compute_baseline_loss 6 other papers
1 dated earlier, 5 later
1708.04782 (2017-08)
StarCraft II: A New Challenge for Reinforcement Learning
1910.03552 (2019-10)
TorchBeast: A PyTorch Platform for Distributed RL
2006.13760 (2020-06)
The NetHack Learning Environment
and 3 more not named here
compute_entropy_loss 5 other papers
5 later
1910.03552 (2019-10)
TorchBeast: A PyTorch Platform for Distributed RL
2006.13760 (2020-06)
The NetHack Learning Environment
2201.03544 (2022-01)
The Effects of Reward Misspecification: Mapping and Mitigating Misaligned Models
and 2 more not named here

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

The reinforcement learning community has made great strides in designing algorithms capable of exceeding human performance on specific tasks. These algorithms are mostly trained one task at the time, each new task requiring to train a brand new agent instance. This means the learning algorithm is general, but each solution is not; each agent can only solve the one task it was trained on. In this work, we study the problem of learning to master not one but multiple sequential-decision tasks at once. A general issue in multi-task learning is that a balance must be found between the needs of multiple tasks competing for the limited resources of a single learning system. Many learning algorithms can get distracted by certain tasks in the set of tasks to solve. Such tasks appear more salient to the learning process, for instance because of the density or magnitude of the in-task rewards. This causes the algorithm to focus on those salient tasks at the expense of generality. We propose to automatically adapt the contribution of each task to the agent's updates, so that all tasks have a similar impact on the learning dynamics. This resulted in state of the art performance on learning to play all games in a set of 57 diverse Atari games. Excitingly, our method learned a single trained policy - with a single set of weights - that exceeds median human performance. To our knowledge, this was the first time a single agent surpassed human-level performance on this multi-task domain. The same approach also demonstrated state of the art performance on a set of 30 tasks in the 3D reinforcement learning platform DeepMind Lab.

For agents

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

get_harvested_code_for_paper("1809.04474")
get_code_for_paper("1809.04474")
have("1809.04474")

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

curl https://syntology.ai/api/ran/1809.04474.json

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

[![Syntology run record](https://syntology.ai/api/ran/1809.04474.svg)](https://syntology.ai/paper/1809.04474)

Connect an agent — have() is free.