SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2306.06327 · 2023

Any-dimensional equivariant neural networks

arXiv · PDF · Open in the Atlas

Code that ran

9 of 12 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 5 have no contract verdict; 4 failures of our instrument; 3 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/2306.06327.json · badge: /api/ran/2306.06327.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
mateodd25/free-nets canonical (classifier label)
classifier basis: census
9 ran · 3 no recorded run
FunctionStatusWhere it lives
LBface_swap Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
mateodd25/free-nets/emlp/datasets.py
code served (permissive licence) · get_code("23ff1cdc894dab43")
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 2, dtype family float, shape (2, 40) 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

# Analyze the function:
# - Takes 'state' parameter
# - Calls state.copy() - so state must be a numpy array
# - Indexes state with [:, L_B_center_top_faces] - so state is 2D
# - L_B_center_top_faces contains indices [33, 25] (4*8+1=33, 3*8+1=25)
# - Maximum index is 33, so second dimension must be at least 34
# - Uses np.roll() on indices

# Create a small state array: batch_size x num_faces
# batch_size=2, num_faces=40 (must be > 33 to accommodate indices)
state = np.arange(2 * 40, dtype=np.float32).reshape(2, 40)

ARGS = {
    "state": state
}
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 2, dtype family float, shape (2, 40) — shape (2, 40) 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
UBedge_flip Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
mateodd25/free-nets/emlp/datasets.py
code served (permissive licence) · get_code("02ec0eac15525baa")
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 2, dtype family float, shape (6, 48) 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

# The function UBedge_flip takes a state parameter
# Looking at the function body:
# - state.copy() is called, so state must be a numpy array
# - state has shape (6, 48) based on the docstring
# - state[:, UB_edge] accesses columns, so it's a 2D array
# - The function performs array indexing and assignment

# Create a state array with shape (6, 48)
state = np.random.randint(0, 2, size=(6, 48)).astype(float)

ARGS = {
    "state": state
}
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 2, dtype family float, shape (6, 48) — shape (6, 48) 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
export Ran
ran; our contract draft was wrong, not the code
mateodd25/free-nets/emlp/utils.py
pointer only (licence: MIT); the stored copy get_code serves, from woriazzc/robust-pu, is pointer only (licence: NONE) · get_code("2660b9d52e1de712")
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 woriazzc/robust-pu, 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.

matrix_power_simple Ran
ran on a synthesized input
mateodd25/free-nets/emlp/groups.py
code served (permissive licence) · get_code("ac065ea6ecd18125")
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 ArrayImpl, ndim 2, dtype family float, shape (4, 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 jax
import jax.numpy as jnp
from jax import jit, vmap

# Create a simple square matrix for matrix power computation
M = jnp.array([
    [2.0, 0.5, 0.1, 0.2],
    [0.3, 1.5, 0.2, 0.1],
    [0.1, 0.2, 1.8, 0.3],
    [0.2, 0.1, 0.4, 2.0]
], dtype=jnp.float32)

# Test with positive integer power
n = 3

ARGS = {
    "M": M,
    "n": n
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, ndim 2, dtype family float, shape (4, 4) — shape (4, 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
noise2sample Ran
ran; our contract draft was wrong, not the code
mateodd25/free-nets/emlp/groups.py
code served (permissive licence) · get_code("8dadbcba29a5fa50")
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 ArrayImpl, ndim 2, dtype family float, shape (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 jax
import jax.numpy as jnp
import numpy as np

# Set seed for reproducibility
np.random.seed(0)

# Define dimensions
D = 4  # dimension of noise vector z
d = 3  # dimension of matrices (d x d)
M = 2  # number of discrete generators
K = 2  # number of repeats for each generator

# z: (D,) - noise vector
z = jnp.array([0.1, 0.2, 0.3, 0.4])

# ks: (M, K) - integer powers for discrete generators
ks = jnp.array([[1, 2], [3, 1]], dtype=jnp.int32)

# lie_algebra: (D, d, d) - basis of Lie algebra
# Create D different (d x d) matrices
lie_algebra = jnp.array([
    [[1.0, 0.0, 0.0],
     [0.0, 0.0, 0.0],
     [0.0, 0.0, 0.0]],
    [[0.0, 1.0, 0.0],
     [0.0, 0.0, 0.0],
     [0.0, 0.0, 0.0]],
    [[0.0, 0.0, 1.0],
     [0.0, 0.0, 0.0],
     [0.0, 0.0, 0.0]],
    [[0.0, 0.0, 0.0],
     [1.0, 0.0, 0.0],
     [0.0, 0.0, 0.0]]
])

# discrete_generators: (M, d, d) - M different (d x d) matrices
discrete_generators = jnp.array([
    [[1.0, 0.1, 0.0],
     [0.0, 1.0, 0.0],
     [0.0, 0.0, 1.0]],
    [[1.0, 0.0, 0.0],
     [0.1, 1.0, 0.0],
     [0.0, 0.0, 1.0]]
])

# seed parameter
seed = 42

ARGS = {
    "z": z,
    "ks": ks,
    "lie_algebra": lie_algebra,
    "discrete_generators": discrete_generators,
    "seed": seed
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, ndim 2, dtype family float, shape (3, 3) — shape (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
random_sample Ran
ran on a synthesized input
mateodd25/free-nets/experiments/free_symmetric_projection.py
code served (permissive licence) · get_code("ad3bd41e9ed1bb19")
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 2, dtype family float, shape (5, 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 numpy as np

def random_sample(size):
    return np.random.randn(size, size)

ARGS = {
    "size": 5
}
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 2, dtype family float, shape (5, 5) — shape (5, 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
random_sample Ran
ran on a synthesized input
mateodd25/free-nets/experiments/aux_O.py
code served (permissive licence) · get_code("5c71362caaf3b0ab")
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 (10,) 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 = {
    "size": 5
}
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 (10,) — shape (10,) 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
rel_err Ran
ran on a synthesized input
mateodd25/free-nets/emlp/groups.py
code served (permissive licence) · get_code("b9e669d60e5fb206")
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 ArrayImpl, 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 jax.numpy as jnp
import numpy as np

# Create two small JAX arrays for testing the relative error function
# Using distinct shapes to make dimensions unambiguous
A = jnp.array([[1.0, 2.0, 3.0, 4.0],
               [5.0, 6.0, 7.0, 8.0]])

B = jnp.array([[1.1, 2.1, 2.9, 4.2],
               [5.2, 5.9, 7.1, 7.8]])

ARGS = {
    "A": A,
    "B": B,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, 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
scale_adjusted_rel_err Ran
ran on a synthesized input
mateodd25/free-nets/experiments/aux_O.py
code served (permissive licence) · get_code("1c0546beb8cb8de2")
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 ArrayImpl, 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 jax.numpy as jnp
import numpy as np

# Construct small valid arrays for the function
a = jnp.array([[1.0, 2.0, 3.0, 4.0],
                [5.0, 6.0, 7.0, 8.0]])

b = jnp.array([[1.1, 2.1, 3.1, 4.1],
                [5.1, 6.1, 7.1, 8.1]])

g = jnp.array([[0.9, 0.1, 0.0, 0.0],
                [0.1, 0.95, 0.05, 0.0],
                [0.0, 0.05, 1.0, 0.0],
                [0.0, 0.0, 0.0, 1.05]])

ARGS = {
    "a": a,
    "b": b,
    "g": g
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
jax, numpy — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type ArrayImpl, 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
ULBcorner_rot Not yet run
unverified (harvested, no recorded run)
mateodd25/free-nets/emlp/datasets.py
code served (permissive licence) · get_code("afeb3056cc5cad08")
How it ran

no invocation recorded.

equivariance_err Not yet run
unverified (harvested, no recorded run)
mateodd25/free-nets/experiments/aux_O.py
code served (permissive licence) · get_code("5fbab6adb8505757")
How it ran

no invocation recorded.

true_mapping Not yet run
unverified (harvested, no recorded run)
mateodd25/free-nets/experiments/free_symmetric_projection.py
code served (permissive licence) · get_code("b0c36eb0a737fbf7")
How it ran

no invocation recorded.

Also present in other papers' code

Of 12 samples lifted from this paper's linked repositories, 12 carry a function identity and 0 do not. Of the 12 distinct functions with an identity, 8 are also present in other papers whose code we hold (8 with at least one paper dated earlier, 0 the same month, 1 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
export 6 other papers
5 dated earlier, 1 later
1904.04717 (2019-04)
Label Propagation for Deep Semi-supervised Learning
2104.09459 (2021-04)
A Practical Method for Constructing Equivariant Multilayer Perceptrons for Arbitrary Matrix Groups
2106.06610 (2021-06)
Scalars are universal: Equivariant machine learning, structured like classical physics
and 3 more not named here
LBface_swap 3 other papers
3 dated earlier
2104.09459 (2021-04)
A Practical Method for Constructing Equivariant Multilayer Perceptrons for Arbitrary Matrix Groups
2106.06610 (2021-06)
Scalars are universal: Equivariant machine learning, structured like classical physics
2204.00887 (2022-04)
Dimensionless machine learning: Imposing exact units equivariance
UBedge_flip 3 other papers
3 dated earlier
2104.09459 (2021-04)
A Practical Method for Constructing Equivariant Multilayer Perceptrons for Arbitrary Matrix Groups
2106.06610 (2021-06)
Scalars are universal: Equivariant machine learning, structured like classical physics
2204.00887 (2022-04)
Dimensionless machine learning: Imposing exact units equivariance
ULBcorner_rot 3 other papers
3 dated earlier
2104.09459 (2021-04)
A Practical Method for Constructing Equivariant Multilayer Perceptrons for Arbitrary Matrix Groups
2106.06610 (2021-06)
Scalars are universal: Equivariant machine learning, structured like classical physics
2204.00887 (2022-04)
Dimensionless machine learning: Imposing exact units equivariance
matrix_power_simple 3 other papers
3 dated earlier
2104.09459 (2021-04)
A Practical Method for Constructing Equivariant Multilayer Perceptrons for Arbitrary Matrix Groups
2106.06610 (2021-06)
Scalars are universal: Equivariant machine learning, structured like classical physics
2204.00887 (2022-04)
Dimensionless machine learning: Imposing exact units equivariance
noise2sample 3 other papers
3 dated earlier
2104.09459 (2021-04)
A Practical Method for Constructing Equivariant Multilayer Perceptrons for Arbitrary Matrix Groups
2106.06610 (2021-06)
Scalars are universal: Equivariant machine learning, structured like classical physics
2204.00887 (2022-04)
Dimensionless machine learning: Imposing exact units equivariance
rel_err 3 other papers
3 dated earlier
2104.09459 (2021-04)
A Practical Method for Constructing Equivariant Multilayer Perceptrons for Arbitrary Matrix Groups
2106.06610 (2021-06)
Scalars are universal: Equivariant machine learning, structured like classical physics
2204.00887 (2022-04)
Dimensionless machine learning: Imposing exact units equivariance
scale_adjusted_rel_err 1 other paper
1 dated earlier
2104.09459 (2021-04)
A Practical Method for Constructing Equivariant Multilayer Perceptrons for Arbitrary Matrix Groups

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

Traditional supervised learning aims to learn an unknown mapping by fitting a function to a set of input-output pairs with a fixed dimension. The fitted function is then defined on inputs of the same dimension. However, in many settings, the unknown mapping takes inputs in any dimension; examples include graph parameters defined on graphs of any size and physics quantities defined on an arbitrary number of particles. We leverage a newly-discovered phenomenon in algebraic topology, called representation stability, to define equivariant neural networks that can be trained with data in a fixed dimension and then extended to accept inputs in any dimension. Our approach is user-friendly, requiring only the network architecture and the groups for equivariance, and can be combined with any training procedure. We provide a simple open-source implementation of our methods and offer preliminary numerical experiments.

For agents

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

get_harvested_code_for_paper("2306.06327")
get_code_for_paper("2306.06327")
have("2306.06327")

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

curl https://syntology.ai/api/ran/2306.06327.json

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

[![Syntology run record](https://syntology.ai/api/ran/2306.06327.svg)](https://syntology.ai/paper/2306.06327)

Connect an agent — have() is free.