SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2404.14064 · 2024

Multi-view Disentanglement for Reinforcement Learning with Multiple Cameras

arXiv · PDF · Open in the Atlas

Code that ran

10 of 11 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; 1 honoured the contract we drafted, 1 violated it and 0 have no contract verdict; 8 failures of our instrument; 1 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/2404.14064.json · badge: /api/ran/2404.14064.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
uoe-agents/mvd canonical (classifier label)
classifier basis: census
10 ran · 1 no recorded run
FunctionStatusWhere it lives
cosine_distance Ran
ran; our contract draft was wrong, not the code
uoe-agents/mvd/algorithms/info_nce.py
code served (permissive licence) · get_code("b5e5ec5ae2bf2ea6")
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, 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 torch
import torch.nn.functional as F

torch.manual_seed(0)

# Create two 2D tensors with the same number of samples and feature dimensions
batch_size = 2
feature_dim = 4

query = torch.randn(batch_size, feature_dim)
positive_key = torch.randn(batch_size, feature_dim)

ARGS = {
    'query': query,
    'positive_key': positive_key,
}
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, 1) — shape (2, 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
full_v1_path_for Ran
ran; our contract draft was wrong, not the code
uoe-agents/mvd/metaworld/envs/asset_path_utils.py
pointer only (licence: MIT); the stored copy get_code serves, from fuyw/TemporalOT, is pointer only (licence: NONE) · get_code("e13df24e0d039aa9")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from fuyw/TemporalOT, and GitHub detected no licence in that repository 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.

full_v2_path_for Ran
ran; our contract draft was wrong, not the code
uoe-agents/mvd/metaworld/envs/asset_path_utils.py
pointer only (licence: MIT); the stored copy get_code serves, from fuyw/TemporalOT, is pointer only (licence: NONE) · get_code("061fee9f0adfe735")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from fuyw/TemporalOT, and GitHub detected no licence in that repository 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.

info_nce Ran
ran; our contract draft was wrong, not the code
uoe-agents/mvd/algorithms/info_nce.py
code served (permissive licence) · get_code("c105e24c6a7befa4")
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)

# Create small tensors with distinct dimensions
batch_size = 2
embedding_dim = 4
num_negatives = 3

# Query tensor: (batch_size, embedding_dim)
query = torch.randn(batch_size, embedding_dim)

# Positive key tensor: (batch_size, embedding_dim)
positive_key = torch.randn(batch_size, embedding_dim)

# Negative keys tensor in 'unpaired' mode: (num_negatives, embedding_dim)
negative_keys = torch.randn(num_negatives, embedding_dim)

# Temperature parameter
temperature = 0.1

# Reduction parameter
reduction = 'mean'

# Negative mode parameter
negative_mode = 'unpaired'

ARGS = {
    'query': query,
    'positive_key': positive_key,
    'negative_keys': negative_keys,
    'temperature': temperature,
    'reduction': reduction,
    'negative_mode': negative_mode
}
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
mlp Ran
ran; our contract draft was wrong, not the code
uoe-agents/mvd/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from xingyu-lin/softagent, is pointer only (licence: NONE) · get_code("bdef24507a4ccfa7")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from xingyu-lin/softagent, and GitHub detected no licence in that repository 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.

preprocess_obs Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
uoe-agents/mvd/utils.py
code served (permissive licence) · get_code("975f23eeea267e55")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, Claude Sonnet 4.5, prompt python_fixture_v1) and returned Tensor, ndim 4, dtype family float, shape (2, 3, 4, 5) 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 image tensor with shape (batch=2, channels=3, height=4, width=5)
# Values should be in [0, 255] range as typical for image preprocessing
obs = torch.randint(0, 256, (2, 3, 4, 5), dtype=torch.float32)

# bits parameter - using default value of 5
bits = 5

ARGS = {
    'obs': obs,
    'bits': bits
}
Setup block drafted by
a language model, Claude Sonnet 4.5, prompt python_fixture_v1
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 4, dtype family float, shape (2, 3, 4, 5) — shape (2, 3, 4, 5) 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
rect_prism_tolerance Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
uoe-agents/mvd/metaworld/envs/reward_utils.py
code served (permissive licence) · get_code("083eee6f7eed5a7d")
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 float64, ndim 0, dtype family float; 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 numpy as np

# Construct valid arguments for rect_prism_tolerance
curr = np.array([0.5, 0.5, 0.5])
zero = np.array([0.0, 0.0, 0.0])
one = np.array([1.0, 1.0, 1.0])

ARGS = {
    "curr": curr,
    "zero": zero,
    "one": one
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float64, ndim 0, dtype family float — no output shape is stated: observe_signature.py recorded float ndim 0, which disagrees with the fixture run
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
to_np Ran
ran, honoured the contract we drafted
uoe-agents/mvd/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from xingyu-lin/softagent, is pointer only (licence: NONE) · get_code("6ee8c9901ab984a9")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (MIT), but get_code serves the stored copy, first harvested from xingyu-lin/softagent, and GitHub detected no licence in that repository 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.

tolerance Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
uoe-agents/mvd/metaworld/envs/reward_utils.py
code served (permissive licence) · get_code("1ab84d50eed1d70e")
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 1, dtype family float, shape (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 numpy as np

ARGS = {
    'x': np.array([0.5, 1.5, -0.3, 2.0]),
    'bounds': (0.0, 1.0),
    'margin': 0.5,
    'sigmoid': 'gaussian',
    'value_at_margin': 0.1
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ndarray, ndim 1, dtype family float, shape (4,) — shape (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
transpose Ran
ran, violated the contract we drafted
uoe-agents/mvd/algorithms/info_nce.py
code served (permissive licence) · get_code("55f1419d4d8c483c")
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 Tensor, ndim 4, dtype family float, shape (2, 4, 5, 3) 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)

# The transpose function takes a single parameter 'x' and calls .transpose(-2, -1) on it
# This requires x to be a tensor with at least 2 dimensions
# We'll create a small tensor with distinct dimensions to make shape probing unambiguous

x = torch.randn(2, 4, 3, 5)  # 4D tensor with distinct sizes

ARGS = {
    "x": x
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1
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 4, dtype family float, shape (2, 4, 5, 3) — shape (2, 4, 5, 3) 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
inverse_tolerance Not yet run
unverified (harvested, no recorded run)
uoe-agents/mvd/metaworld/envs/reward_utils.py
code served (permissive licence) · get_code("e1dac90f131811f7")
How it ran

no invocation recorded.

Also present in other papers' code

Of 11 samples lifted from this paper's linked repositories, 11 carry a function identity and 0 do not. Of the 11 distinct functions with an identity, 7 are also present in other papers whose code we hold (7 with at least one paper dated earlier, 0 the same month, 6 later) and 4 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
transpose 17 other papers
8 dated earlier, 8 later, 1 same year, order not determinable · 2 only as an assembly brick
1910.12037 (2019-10)
Region Mutual Information Loss for Semantic Segmentation
2207.05500 (2022-07)
Modality-Aware Contrastive Instance Learning with Self-Distillation for Weakly-Supervised Audio-Visual Violence Detection
2306.10008 (2023-06)
CLIP2Protect: Protecting Facial Privacy using Text-Guided Makeup via Adversarial Latent Search
and 14 more not named here
mlp 11 other papers
8 dated earlier, 3 later · 2 only as an assembly brick
1812.05905 (2018-12)
Soft Actor-Critic Algorithms and Applications
2004.13649 (2020-04)
Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixels
2110.04935 (2021-10)
Learning Temporally-Consistent Representations for Data-Efficient Reinforcement Learning
and 8 more not named here
to_np 9 other papers
5 dated earlier, 4 later · 2 only as an assembly brick
1812.05905 (2018-12)
Soft Actor-Critic Algorithms and Applications
2004.13649 (2020-04)
Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixels
2110.04935 (2021-10)
Learning Temporally-Consistent Representations for Data-Efficient Reinforcement Learning
and 6 more not named here
preprocess_obs 4 other papers
4 dated earlier
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
2010.09163 (2020-10)
D2RL: Deep Dense Architectures in Reinforcement Learning
and 1 more not named here
full_v1_path_for 2 other papers
1 dated earlier, 1 later
1910.10897 (2019-10)
Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learning
2410.21795 (2024-10)
Robot Policy Learning with Temporal Optimal Transport Reward
full_v2_path_for 2 other papers
1 dated earlier, 1 later
1910.10897 (2019-10)
Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learning
2410.21795 (2024-10)
Robot Policy Learning with Temporal Optimal Transport Reward
inverse_tolerance 2 other papers
1 dated earlier, 1 later
1910.10897 (2019-10)
Meta-World: A Benchmark and Evaluation for Multi-Task and Meta Reinforcement Learning
2410.21795 (2024-10)
Robot Policy Learning with Temporal Optimal Transport Reward

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 performance of image-based Reinforcement Learning (RL) agents can vary depending on the position of the camera used to capture the images. Training on multiple cameras simultaneously, including a first-person egocentric camera, can leverage information from different camera perspectives to improve the performance of RL. However, hardware constraints may limit the availability of multiple cameras in real-world deployment. Additionally, cameras may become damaged in the real-world preventing access to all cameras that were used during training. To overcome these hardware constraints, we propose Multi-View Disentanglement (MVD), which uses multiple cameras to learn a policy that is robust to a reduction in the number of cameras to generalise to any single camera from the training set. Our approach is a self-supervised auxiliary task for RL that learns a disentangled representation from multiple cameras, with a shared representation that is aligned across all cameras to allow generalisation to a single camera, and a private representation that is camera-specific. We show experimentally that an RL agent trained on a single third-person camera is unable to learn an optimal policy in many control tasks; but, our approach, benefiting from multiple cameras during training, is able to solve the task using only the same single third-person camera.

For agents

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

get_harvested_code_for_paper("2404.14064")
get_code_for_paper("2404.14064")
have("2404.14064")

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

curl https://syntology.ai/api/ran/2404.14064.json

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

[![Syntology run record](https://syntology.ai/api/ran/2404.14064.svg)](https://syntology.ai/paper/2404.14064)

Connect an agent — have() is free.