SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2410.23595 · NeurIPS · 2024

Disentangling Interpretable Factors with Supervised Independent Subspace Principal Component Analysis

Jiayu Su, David Knowles, Raul Rabadan

arXiv · PDF · Open in the Atlas

Code that ran

13 of 18 samples (functions and classes) lifted from this paper's linked repositories executed on a synthesized input in Syntology's jail (2 of them constructed an object rather than computing a result; 1 honoured the contract we drafted, 0 violated it and 3 have no contract verdict; 9 failures of our instrument; 5 no recorded run), measured 2026-11-22; 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/2410.23595.json · badge: /api/ran/2410.23595.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
nitzanlab/biolord_reproducibility canonical (classifier label)
classifier basis: census
4 ran · 1 no recorded run
JiayuSuPKU/sispca marked official in Papers with Code
Papers with Code archive
9 ran · 4 no recorded run
FunctionStatusWhere it lives
Kernel Ran
ran on a synthesized input · constructed an object; no result was computed
JiayuSuPKU/sispca/sispca/model.py
code served (permissive licence) · get_code("0f609a43f773f0b5")
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 an instance of its own class, Kernel: the run constructed the object and did not call it, so no result was computed. 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
import numpy as np

torch.manual_seed(0)

# Create a small Q matrix for 'continuous' type
Q_continuous = torch.randn(2, 4)

# Create a small identity dimension
identity_dim = 3

# Create a small custom kernel matrix
custom_kernel = torch.randn(2, 2)
custom_kernel = (custom_kernel + custom_kernel.T) / 2  # Make it symmetric

ARGS = {
    "target_type": "continuous",
    "Q": Q_continuous,
    "target_kernel": None,
}
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 Kernel — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a Kernel 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
LossHistoryCallback Ran
ran on a synthesized input
JiayuSuPKU/sispca/sispca/model.py
code served (permissive licence) · get_code("b950bfa4d8560e86")
How it ran

Ran in Syntology's sandbox on this synthesized input (drafted by a language model, class construct-then-call prompt, model not recorded) and returned NoneType. 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 lightning as L
import torch
import torch.nn as nn

class LossHistoryCallback(L.Callback):

    def __init__(self):
        super().__init__()
        self.loss_history = {'train_loss_step': [], 'train_loss_epoch': [], 'reg_loss_step': [], 'reg_loss_epoch': []}

    def on_train_batch_end(self, trainer, pl_module, outputs, batch, batch_idx):
        self.loss_history['train_loss_step'].append(outputs['train_loss'].item())
        self.loss_history['reg_loss_step'].append(outputs['reg_loss'].item())

    def on_train_epoch_end(self, trainer, pl_module):
        metrics = trainer.callback_metrics
        self.loss_history['train_loss_epoch'].append(metrics['train_loss_epoch'].item())
        self.loss_history['reg_loss_epoch'].append(metrics['reg_loss_epoch'].item())


INIT_ARGS = {}

ARGS = {
    'trainer': type('MockTrainer', (), {
        'callback_metrics': {
            'train_loss_epoch': torch.tensor(0.5),
            'reg_loss_epoch': torch.tensor(0.1)
        }
    })(),
    'pl_module': type('MockModule', (), {})(),
    'outputs': {
        'train_loss': torch.tensor(0.45),
        'reg_loss': torch.tensor(0.09)
    },
    'batch': torch.randn(2, 4),
    'batch_idx': 0
}

METHOD = 'on_train_batch_end'

callback = LossHistoryCallback()
callback.on_train_batch_end(**ARGS)
callback.on_train_epoch_end(ARGS['trainer'], ARGS['pl_module'])
Setup block drafted by
a language model, class construct-then-call prompt, model not recorded
Imports
lightning, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type NoneType — a NoneType output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a NoneType 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: torch and numpy seeded to 0 by the harness immediately before the class was constructed from INIT_ARGS (fixture_check.py)
PCA Ran
ran on a synthesized input · constructed an object; no result was computed
JiayuSuPKU/sispca/sispca/model.py
code served (permissive licence) · get_code("87d40b6c10e88350")
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 an instance of its own class, PCA: the run constructed the object and did not call it, so no result was computed. 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
import torch.nn as nn
import lightning as L

torch.manual_seed(0)

ARGS = {
    'n_feature': 4,
    'n_latent': 3,
    'lr': 0.5,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
lightning, torch — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type PCA — the run constructed an instance of its own class and did not call it; no shape is stated — the instrument coerces any output to an array; for a PCA 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
bool2idx Ran
ran, honoured the contract we drafted
nitzanlab/biolord_reproducibility/utils/utils_perturbations.py
code served (permissive licence) · get_code("1feac17484b565e2")
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 int, shape (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 numpy as np

# Create a small boolean array with some True and False values
x = np.array([True, False, True, False, True])

ARGS = {
    "x": x
}
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 int, shape (3,) — shape (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
compute_r2 Ran
ran; our contract draft was wrong, not the code
nitzanlab/biolord_reproducibility/utils/utils_perturbations_sciplex3.py
code served (permissive licence) · get_code("bdb2cc41a701eff6")
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 float, 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
from torchmetrics import R2Score

torch.manual_seed(0)

# Create small tensors for y_true and y_pred
# y_true: ground truth labels, shape (2, 4)
y_true = torch.randn(2, 4)

# y_pred: predictions, shape (2, 4)
y_pred = torch.randn(2, 4)

ARGS = {
    "y_true": y_true,
    "y_pred": y_pred,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
torch, torchmetrics — import names as written, not package names to install; a name the repository itself provides is not a package
Returned
type float, 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
gaussian_kernel Ran
ran; our contract draft was wrong, not the code
JiayuSuPKU/sispca/sispca/model.py
code served (permissive licence) · get_code("1f9ef96229ba4b75")
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, 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 2D tensor with shape (n_sample, n_feature)
# Using distinct dimensions: 2 samples, 4 features
x = torch.randn(2, 4)

# bw parameter can be None (will be computed as median distance)
bw = None

ARGS = {
    "x": x,
    "bw": bw,
}
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, 2) — shape (2, 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
gram_schmidt Ran
ran; our contract draft was wrong, not the code
JiayuSuPKU/sispca/sispca/model.py
code served (permissive licence) · get_code("9b25806574e96a9e")
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 (4, 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 a small 2D tensor with orthogonal structure
# Using a tall-and-thin matrix (more rows than columns) for QR decomposition
x = torch.randn(4, 3, dtype=torch.float32)

ARGS = {
    'x': x
}
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 (4, 3) — shape (4, 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
hsic_gaussian Ran
ran; our contract draft was wrong, not the code
JiayuSuPKU/sispca/sispca/model.py
code served (permissive licence) · get_code("c522d8da10a5aa0a")
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 lightning as L

torch.manual_seed(0)

# Create small 2D tensors
# x: (n_sample=2, n_feature_1=4)
# y: (n_sample=2, n_feature_2=3)
x = torch.randn(2, 4)
y = torch.randn(2, 3)

# bw can be None (uses median distance) or a specific value
bw = None

ARGS = {
    "x": x,
    "y": y,
    "bw": bw,
}
Setup block drafted by
a language model, Claude Haiku 4.5, prompt python_fixture_v1, through the batch API
Imports
lightning, 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
hsic_linear Ran
ran; our contract draft was wrong, not the code
JiayuSuPKU/sispca/sispca/model.py
code served (permissive licence) · get_code("568126fd0824d1ac")
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

torch.manual_seed(0)

# Create two 2D tensors with distinct dimensions
# x: (n_sample=2, n_feature_1=4)
# y: (n_sample=2, n_feature_2=3)
x = torch.randn(2, 4)
y = torch.randn(2, 3)

ARGS = {
    "x": x,
    "y": y,
}
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
normalize_col Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
JiayuSuPKU/sispca/sispca/model.py
code served (permissive licence) · get_code("92ec5c1f2e7e0128")
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 2D tensor (n_sample=2, n_feature=4)
x = torch.randn(2, 4)

ARGS = {
    "x": x,
    "center": True,
    "scale": 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 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
repeat_n Ran
ran; our contract draft was wrong, not the code
nitzanlab/biolord_reproducibility/utils/utils_perturbations.py
code served (permissive licence) · get_code("e02fc89779a42341")
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 (5, 12) as a later run recorded it. The image it ran in was not recorded. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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 test tensor
# The function expects a tensor x and will reshape it to (1, -1) then repeat along axis 0
# Let's create a 2D tensor with distinct dimensions for clarity
x = torch.randn(3, 4)  # 3x4 tensor

# The repetition count
n = 5

ARGS = {
    'x': x,
    'n': n
}
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 (5, 12) — shape (5, 12) 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
repeat_n Ran
ran; our contract draft was wrong, not the code
nitzanlab/biolord_reproducibility/utils/utils_perturbations_sciplex3.py
code served (permissive licence) · get_code("8633543f3eee6944")
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. The code mentions cuda, and the harness aliases cuda to the CPU for every sample; whether this run needed that is 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
# The function expects a tensor x and will reshape it to (1, -1) then repeat along axis 0
# Let's create a tensor with shape (4,) so it becomes (1, 4) after view, then (n, 4) after repeat
x = torch.randn(4)

# Create the repetition count
n = 3

ARGS = {
    'x': x,
    'n': n,
}
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
tr_cov Ran
ran; our contract draft was wrong, not the code
JiayuSuPKU/sispca/sispca/model.py
code served (permissive licence) · get_code("9e6788e5b56a53c4")
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

torch.manual_seed(0)

# Construct a small 2D tensor (n_sample=2, n_feature=4)
x = torch.randn(2, 4)

ARGS = {
    "x": x
}
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
SISPCA Not yet run
unverified (harvested, no recorded run)
JiayuSuPKU/sispca/sispca/model.py
code served (permissive licence) · get_code("9e6e893020f51961")
How it ran

no invocation recorded.

SISPCADataset Not yet run
unverified (harvested, no recorded run)
JiayuSuPKU/sispca/sispca/model.py
code served (permissive licence) · get_code("6589b8d5c44e13b8")
How it ran

no invocation recorded.

Supervision Not yet run
unverified (harvested, no recorded run)
JiayuSuPKU/sispca/sispca/model.py
code served (permissive licence) · get_code("79ea314871c66c92")
How it ran

no invocation recorded.

compute_metrics Not yet run
unverified (harvested, no recorded run)
nitzanlab/biolord_reproducibility/utils/utils_perturbations.py
code served (permissive licence) · get_code("7a8d80823166d9e6")
How it ran

no invocation recorded.

slice_sparse_matrix Not yet run
unverified (harvested, no recorded run)
JiayuSuPKU/sispca/sispca/model.py
code served (permissive licence) · get_code("5e4292b883636890")
How it ran

no invocation recorded.

Also present in other papers' code

Of 18 samples lifted from this paper's linked repositories, 18 carry a function identity and 0 do not. Of the 17 distinct functions with an identity, 0 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 0 later) and 17 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.

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 success of machine learning models relies heavily on effectively representing high-dimensional data. However, ensuring data representations capture humanunderstandable concepts remains difficult, often requiring the incorporation of prior knowledge and decomposition of data into multiple subspaces. Traditional linear methods fall short in modeling more than one space, while more expressive deep learning approaches lack interpretability. Here, we introduce Supervised Independent Subspace Principal Component Analysis (sisPCA), a PCA extension designed for multi-subspace learning. Leveraging the Hilbert-Schmidt Independence Criterion (HSIC), sisPCA incorporates supervision and simultaneously ensures subspace disentanglement. We demonstrate sisPCA's connections with autoencoders and regularized linear regression and showcase its ability to identify and separate hidden data structures through extensive applications, including breast cancer diagnosis from image features, learning aging-associated DNA methylation changes, and single-cell analysis of malaria infection. Our results reveal distinct functional pathways associated with malaria colonization, underscoring the essentiality of explainable representation in high-dimensional data analysis.

For agents

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

get_harvested_code_for_paper("2410.23595")
get_code_for_paper("2410.23595")
have("2410.23595")

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

curl https://syntology.ai/api/ran/2410.23595.json

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

[![Syntology run record](https://syntology.ai/api/ran/2410.23595.svg)](https://syntology.ai/paper/2410.23595)

Connect an agent — have() is free.