SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2202.07559 · 2022

Unsupervised Learning of Group Invariant and Equivariant Representations

arXiv · PDF · Open in the Atlas

Code that ran

12 of 16 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 9 have no contract verdict; 3 failures of our instrument; 4 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/2202.07559.json · badge: /api/ran/2202.07559.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
jrwnter/group-invaraint-ae canonical (classifier label)
classifier basis not recorded on the link
12 ran · 4 no recorded run
FunctionStatusWhere it lives
add_arguments Ran
ran on a synthesized input
jrwnter/group-invaraint-ae/giae/se3/hyperparameter.py
code served (permissive licence) · get_code("801c9be8f9ea49be")
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 ArgumentParser. 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 argparse
import os

torch_imported = False
try:
    import torch
    torch.manual_seed(0)
    torch_imported = True
except ImportError:
    pass

DEFAULT_SAVE_DIR = os.path.join(os.getcwd(), 'saves0')

ARGS = {
    'parser': argparse.ArgumentParser()
}
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 ArgumentParser — a ArgumentParser output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ArgumentParser 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
add_arguments Ran
ran on a synthesized input
jrwnter/group-invaraint-ae/giae/sn/hyperparameter.py
code served (permissive licence) · get_code("7779401973a8b2ff")
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 ArgumentParser. 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 os
import argparse
import torch

torch.manual_seed(0)

DEFAULT_SAVE_DIR = os.path.join(os.getcwd(), 'saves')

ARGS = {
    'parser': argparse.ArgumentParser()
}
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 ArgumentParser — a ArgumentParser output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a ArgumentParser 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
angles_to_matrix Ran
ran; our contract draft was wrong, not the code
jrwnter/group-invaraint-ae/giae/se3/data.py
code served (permissive licence) · get_code("48f03a318b285378")
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 4, dtype family float, shape (2, 4, 3, 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)

# Create small tensor arguments for angles_to_matrix
# The function expects three tensors of the same shape (or broadcastable)
# Each represents rotation angles in radians
alpha = torch.randn(2, 4)
beta = torch.randn(2, 4)
gamma = torch.randn(2, 4)

ARGS = {
    "alpha": alpha,
    "beta": beta,
    "gamma": gamma,
}
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 4, dtype family float, shape (2, 4, 3, 3) — shape (2, 4, 3, 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
get_node_mask_batch Ran
ran on a synthesized input
jrwnter/group-invaraint-ae/giae/se3/modules.py
code served (permissive licence) · get_code("a4854c0652f65b0b")
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 (3, 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)

# Construct arguments for get_node_mask_batch
# batch: Tensor of node assignments to graphs
# batch_num_nodes: Tensor containing number of nodes per graph
# max_num_nodes: int, maximum nodes in any graph
# batch_size: Optional[int], number of graphs

batch = torch.tensor([0, 0, 1, 1, 1, 2], dtype=torch.long)
batch_num_nodes = torch.tensor([2, 3, 1], dtype=torch.long)
max_num_nodes = 4
batch_size = 3

ARGS = {
    "batch": batch,
    "batch_num_nodes": batch_num_nodes,
    "max_num_nodes": max_num_nodes,
    "batch_size": batch_size
}
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 (3, 4) — shape (3, 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
get_rotation_matrix Ran
ran on a synthesized input
jrwnter/group-invaraint-ae/giae/so2/modules.py
code served (permissive licence) · get_code("fc9401546b6c29b3")
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 3, dtype family float, shape (2, 2, 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)

# Analyze the function:
# - v is a tensor that gets normalized
# - v is indexed with [:, 0] and [:, 1], so it needs shape (batch_size, 2) at minimum
# - The function creates rotation matrices, using v.size(0) for batch dimension
# - Output rot has shape (batch_size, 2, 2, 3) based on the stack operations
# We'll create a batch of 2 vectors, each with 2 components

v = torch.randn(2, 2)

ARGS = {
    "v": v,
    "eps": 0.0001
}
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 3, dtype family float, shape (2, 2, 3) — shape (2, 2, 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
get_rotation_matrix_from_two_vector Ran
ran on a synthesized input
jrwnter/group-invaraint-ae/giae/se3/utils.py
code served (permissive licence) · get_code("745a9ffc444e2eb2")
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 3, dtype family float, shape (2, 3, 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)

# Create two random 3D vectors in a batch of 2
# The function expects v1 and v2 to be tensors of shape (batch_size, 3)
# since it uses torch.cross(v1, v2, dim=-1) which requires 3D vectors
v1 = torch.randn(2, 3)
v2 = torch.randn(2, 3)

ARGS = {
    "v1": v1,
    "v2": v2,
}
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 3, dtype family float, shape (2, 3, 3) — shape (2, 3, 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
matrix_x Ran
ran; our contract draft was wrong, not the code
jrwnter/group-invaraint-ae/giae/se3/data.py
code served (permissive licence) · get_code("7ecf56522713a0e7")
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 5, dtype family float, shape (2, 2, 4, 3, 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 function takes an angle tensor of any shape and computes rotation matrices
# It uses .cos(), .sin(), ones_like(), zeros_like(), and stack operations
# So we need a tensor of angles with arbitrary shape

angle = torch.tensor([
    [[0.1, 0.2, 0.3, 0.4],
     [0.5, 0.6, 0.7, 0.8]],
    [[0.9, 1.0, 1.1, 1.2],
     [1.3, 1.4, 1.5, 1.6]]
])

ARGS = {
    "angle": angle
}
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 5, dtype family float, shape (2, 2, 4, 3, 3) — shape (2, 2, 4, 3, 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
norm_no_nan Ran
ran on a synthesized input
jrwnter/group-invaraint-ae/giae/se3/modules.py
code served (permissive licence) · get_code("2e0f13acc894019d")
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 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 torch

torch.manual_seed(0)

# Create a small tensor for testing
# Using shape (2, 4) - batch size 2, dimension 4
x = torch.randn(2, 4)

ARGS = {
    "x": x,
    "axis": -1,
    "keepdims": False,
    "eps": 1e-08,
    "sqrt": True,
}
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 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
normalize_vector Ran
ran on a synthesized input
jrwnter/group-invaraint-ae/giae/se3/utils.py
code served (permissive licence) · get_code("0424a9c044540d5c")
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 vector tensor with shape (2, 4)
# dim=-1 corresponds to the last dimension, so we normalize along dimension 1
v = torch.randn(2, 4)

ARGS = {
    "v": v,
    "eps": 1e-08,
}
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
orthogonal_projection Ran
ran on a synthesized input
jrwnter/group-invaraint-ae/giae/se3/utils.py
code served (permissive licence) · get_code("2c04bd41c8fbf281")
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)

# The function orthogonal_projection takes two batched vectors v1 and v2
# and projects v2 to be orthogonal to v1 using the formula:
# v2 = v2 - (v1 · v2) * v1

# v1 and v2 should be tensors of shape (batch_size, dim)
# The function uses:
# - unsqueeze() to reshape for bmm
# - bmm() for batch matrix multiplication
# - squeeze() to remove dimensions
# - unsqueeze(-1) to add dimensions back
# - element-wise multiplication with *

batch_size = 2
dim = 4

v1 = torch.randn(batch_size, dim)
v2 = torch.randn(batch_size, dim)

ARGS = {
    "v1": v1,
    "v2": v2,
}
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
rot_img Ran
ran; our contract draft was wrong, not the code
jrwnter/group-invaraint-ae/giae/so2/modules.py
pointer only (licence: BSD-3-Clause); the stored copy get_code serves, from yegonkim/partial_equiv, is pointer only (licence: NONE) · get_code("f18eeb0842879ad9")
How it ran

The body of this sample is withheld: this paper's link allows the body inline (BSD-3-Clause), but get_code serves the stored copy, first harvested from yegonkim/partial_equiv, 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.

sparse_sort Ran
ran on a synthesized input
jrwnter/group-invaraint-ae/giae/se3/modules.py
code served (permissive licence) · get_code("d53cae1b85e25f9d")
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. 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 torch
from torch import Tensor

torch.manual_seed(0)

# Create small tensors for testing
src = torch.randn(2, 4)  # batch=2, dim=4
index = torch.randint(0, 10, (2, 4)).float()  # same shape as src

ARGS = {
    "src": src,
    "index": index,
    "dim": 0,
    "descending": False,
    "eps": 1e-12,
    "dtype": torch.float32
}
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 tuple — a tuple output has no array shape of its own; none is 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
generate_dataset Not yet run
unverified (harvested, no recorded run)
jrwnter/group-invaraint-ae/giae/sn/data.py
code served (permissive licence) · get_code("d138abad047119a8")
How it ran

no invocation recorded.

generate_or_open_data Not yet run
unverified (harvested, no recorded run)
jrwnter/group-invaraint-ae/giae/sn/data.py
code served (permissive licence) · get_code("2b6d9ba0714ccd52")
How it ran

no invocation recorded.

get_non_linearity Not yet run
unverified (harvested, no recorded run)
jrwnter/group-invaraint-ae/giae/so2/modules.py
code served (permissive licence) · get_code("13aa7ec54fbcac9a")
How it ran

no invocation recorded.

matrix_y Not yet run
unverified (harvested, no recorded run)
jrwnter/group-invaraint-ae/giae/se3/data.py
code served (permissive licence) · get_code("b0d62ee795b41cdb")
How it ran

no invocation recorded.

Also present in other papers' code

Of 16 samples lifted from this paper's linked repositories, 16 carry a function identity and 0 do not. Of the 16 distinct functions with an identity, 5 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 5 later) and 11 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
angles_to_matrix 1 other paper
1 later
2504.06231 (2025-04)
Orb-v3: atomistic simulation at scale
get_non_linearity 1 other paper
1 later
2312.12223 (2023-12)
Self-Supervised Detection of Perfect and Partial Input-Dependent Symmetries
matrix_x 1 other paper
1 later
2504.06231 (2025-04)
Orb-v3: atomistic simulation at scale
matrix_y 1 other paper
1 later
2504.06231 (2025-04)
Orb-v3: atomistic simulation at scale
rot_img 1 other paper
1 later
2407.04271 (2024-07)
Variational Partial Group Convolutions for Input-Aware Partial Equivariance of Rotations and Color-Shifts

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

Equivariant neural networks, whose hidden features transform according to representations of a group G acting on the data, exhibit training efficiency and an improved generalisation performance. In this work, we extend group invariant and equivariant representation learning to the field of unsupervised deep learning. We propose a general learning strategy based on an encoder-decoder framework in which the latent representation is separated in an invariant term and an equivariant group action component. The key idea is that the network learns to encode and decode data to and from a group-invariant representation by additionally learning to predict the appropriate group action to align input and output pose to solve the reconstruction task. We derive the necessary conditions on the equivariant encoder, and we present a construction valid for any G, both discrete and continuous. We describe explicitly our construction for rotations, translations and permutations. We test the validity and the robustness of our approach in a variety of experiments with diverse data types employing different network architectures.

For agents

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

get_harvested_code_for_paper("2202.07559")
get_code_for_paper("2202.07559")
have("2202.07559")

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

curl https://syntology.ai/api/ran/2202.07559.json

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

[![Syntology run record](https://syntology.ai/api/ran/2202.07559.svg)](https://syntology.ai/paper/2202.07559)

Connect an agent — have() is free.