SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2306.09459 · 2023

Recurrent Action Transformer with Memory

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; 2 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; 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/2306.09459.json · badge: /api/ran/2306.09459.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
airi-institute/rate canonical (classifier label)
classifier basis: classifier_2025
3 ran · 0 no recorded run
FunctionStatusWhere it lives
check_atari_env Ran
ran, honoured the contract we drafted
airi-institute/rate/Atari/train_rate_atari.py
code served (permissive licence) · get_code("1328231942c456af")
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 tuple, ndim 1, dtype family int; no output shape is stated. The image it ran in was not recorded. If your copy returns the same type and ndim 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 numpy as np

class MockALE:
    """Mock ALE object with lives() method."""
    def lives(self):
        return 5

class MockAtariEnv:
    """Mock Atari environment with required attributes and methods."""
    def __init__(self):
        self.ale = MockALE()
        self._action_meanings = ['NOOP', 'FIRE', 'RIGHT', 'LEFT']
    
    def get_action_meanings(self):
        return self._action_meanings
    
    def unwrapped(self):
        return self
    
    @property
    def unwrapped(self):
        return self

# Create a valid Atari environment mock
env = MockAtariEnv()

ARGS = {
    'env': env
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type tuple, ndim 1, dtype family int — a tuple output has no array shape of its own; none is stated
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
take_fire_action Ran
ran, honoured the contract we drafted
airi-institute/rate/Atari/train_rate_atari.py
code served (permissive licence) · get_code("8d6bc348195b6ab4")
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 ndarray, ndim 3, dtype family int, shape (84, 84, 1) 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 gym
import numpy as np

class MockAtari(gym.Env):
    """Mock Atari environment for testing."""
    
    def __init__(self):
        self.observation_space = gym.spaces.Box(low=0, high=255, shape=(84, 84, 1), dtype=np.uint8)
        self.action_space = gym.spaces.Discrete(4)
        self._step_count = 0
    
    def reset(self):
        return np.zeros((84, 84, 1), dtype=np.uint8)
    
    def step(self, action):
        """Execute one step of the environment."""
        self._step_count += 1
        observation = np.random.randint(0, 256, (84, 84, 1), dtype=np.uint8)
        reward = 1.0
        done = False
        info = {}
        return observation, reward, done, info
    
    def get_action_meanings(self):
        """Return action meanings."""
        return ['NOOP', 'FIRE', 'UP', 'DOWN']
    
    def unwrapped(self):
        """Return the unwrapped environment."""
        return self

# Create a wrapper that provides unwrapped property
class EnvironmentWrapper:
    def __init__(self, env):
        self.env = env
    
    def step(self, action):
        return self.env.step(action)
    
    @property
    def unwrapped(self):
        return self.env

env = EnvironmentWrapper(MockAtari())

ARGS = {
    'env': env
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
gym, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 3, dtype family int, shape (84, 84, 1) — shape (84, 84, 1) 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
top_k_logits Ran
ran; our contract draft was wrong, not the code
airi-institute/rate/Atari/train_rate_atari.py
code served (permissive licence) · get_code("0f0021cdea13e4da")
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 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 torch

torch.manual_seed(0)

# Create a small logits tensor with shape (batch_size=2, vocab_size=4)
logits = torch.randn(2, 4)

# Set k to a small value less than the vocabulary dimension
k = 2

ARGS = {
    "logits": logits,
    "k": k,
}
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 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

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, 1 are also present in other papers whose code we hold (1 with at least one paper dated earlier, 0 the same month, 1 later) and 2 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
top_k_logits 16 other papers
7 dated earlier, 9 later
2104.07652 (2021-04)
Geometry-Free View Synthesis: Transformers and no 3D Priors
2106.02039 (2021-06)
Offline Reinforcement Learning as One Big Sequence Modeling Problem
2106.14131 (2021-06)
SymbolicGPT: A Generative Transformer Model for Symbolic Regression
and 13 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 2026-11-22.

Repositories linked to this paper

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

Abstract

Transformers have become increasingly popular in offline reinforcement learning (RL) due to their ability to treat agent trajectories as sequences, reframing policy learning as a sequence modeling task. However, in partially observable environments (POMDPs), effective decision-making depends on retaining information about past events -- something that standard transformers struggle with due to the quadratic complexity of self-attention, which limits their context length. One solution to this problem is to extend transformers with memory mechanisms. We propose the Recurrent Action Transformer with Memory (RATE), a novel transformer-based architecture for offline RL that incorporates a recurrent memory mechanism designed to regulate information retention. We evaluate RATE across a diverse set of environments: memory-intensive tasks (ViZDoom-Two-Colors, T-Maze, Memory Maze, Minigrid-Memory, and POPGym), as well as standard Atari and MuJoCo benchmarks. Our comprehensive experiments demonstrate that RATE significantly improves performance in memory-dependent settings while remaining competitive on standard tasks across a broad range of baselines. These findings underscore the pivotal role of integrated memory mechanisms in offline RL and establish RATE as a unified, high-capacity architecture for effective decision-making over extended horizons. Code: https://sites.google.com/view/rate-model/.

For agents

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

get_harvested_code_for_paper("2306.09459")
get_code_for_paper("2306.09459")
have("2306.09459")

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

curl https://syntology.ai/api/ran/2306.09459.json

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

[![Syntology run record](https://syntology.ai/api/ran/2306.09459.svg)](https://syntology.ai/paper/2306.09459)

Connect an agent — have() is free.