SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2206.01323 · NeurIPS · 2022

SPD domain-specific batch normalization to crack interpretable unsupervised domain adaptation in EEG

Jun-Ichiro Hirayama, Reinmar Kobler, Qibin Zhao, Motoaki Kawanabe

arXiv · PDF · Open in the Atlas

Code that ran

5 of 8 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, 1 violated it and 2 have no contract verdict; 2 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/2206.01323.json · badge: /api/ran/2206.01323.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
rkobler/TSMNet canonical (classifier label)
classifier basis not recorded on the link
5 ran · 3 no recorded run
FunctionStatusWhere it lives
broadcast_dims Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
rkobler/TSMNet/spdnets/functionals.py
pointer only (licence: BSD-3-Clause); the stored copy get_code serves, from GitZH-Chen/RMLR, is pointer only (licence: NONE) · get_code("943ae75cff49f97d")
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 GitZH-Chen/RMLR, 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.

ensure_sym Ran
ran, violated the contract we drafted
rkobler/TSMNet/spdnets/functionals.py
pointer only (licence: BSD-3-Clause); the stored copy get_code serves, from gitzh-chen/rmlr, is pointer only (licence: NONE) · get_code("58a1c96ea5709535")
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 gitzh-chen/rmlr, 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.

orderedeigh Ran
ran on a synthesized input
rkobler/TSMNet/experiments/figure3.py
code served (permissive licence) · get_code("c20a253f4ac0abbc")
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
import numpy as np
from scipy.linalg import eigh

torch.manual_seed(0)

# Create small symmetric positive definite matrices for A
# A should have shape (..., n, n) where ... can be multiple batch dimensions
# Based on the code: Aflat = A.flatten(end_dim=-3) suggests A can have arbitrary batch dims
# Let's create A with shape (2, 3, 4, 4) - batch dims (2, 3), matrix dims (4, 4)
A_np = np.random.randn(2, 3, 4, 4)
A_np = (A_np + A_np.transpose(0, 1, 3, 2)) / 2  # Make symmetric
A_np = A_np + np.eye(4) * 5  # Make positive definite
A = torch.tensor(A_np, dtype=torch.float32)

# Create B with shape (1, 4, 4) as required by the assert
# B should also be symmetric positive definite for eigh
B_np = np.random.randn(1, 4, 4)
B_np = (B_np + B_np.transpose(0, 2, 1)) / 2  # Make symmetric
B_np = B_np + np.eye(4) * 5  # Make positive definite
B = torch.tensor(B_np, dtype=torch.float32)

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
numpy, scipy, 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
sum_bcastdims Ran
ran; the contract check could not run (RAISES, DEP_MISSING or TIMEOUT)
rkobler/TSMNet/spdnets/functionals.py
pointer only (licence: BSD-3-Clause); the stored copy get_code serves, from GitZH-Chen/RMLR, is pointer only (licence: NONE) · get_code("1e2aa95942f11e42")
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 GitZH-Chen/RMLR, 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.

summarize_score Ran
ran on a synthesized input
rkobler/TSMNet/experiments/tables.py
code served (permissive licence) · get_code("4d5814f5caeb8534")
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 str. 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)

# Analyze the function: it takes a single parameter 'x' which is a dict
# The dict must have keys: 'pval', 'mean', 'std'
# All three values are used in numeric operations (comparison, multiplication, formatting)

x = {
    'pval': 0.0005,  # Will trigger the most significant decorator
    'mean': 0.075,   # Will be multiplied by 100 for display
    'std': 0.012     # Will be multiplied by 100 for display
}

ARGS = {
    'x': x
}
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 str — a str output has no array shape of its own; none is stated — the instrument coerces any output to an array; for a str 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
difference_within_evaluation Not yet run
unverified (harvested, no recorded run)
rkobler/TSMNet/experiments/resultio.py
code served (permissive licence) · get_code("2427dab9c001d778")
How it ran

no invocation recorded.

loadresults Not yet run
unverified (harvested, no recorded run)
rkobler/TSMNet/experiments/resultio.py
code served (permissive licence) · get_code("ee737b09bff04e28")
How it ran

no invocation recorded.

summarize_within_evaluation Not yet run
unverified (harvested, no recorded run)
rkobler/TSMNet/experiments/resultio.py
code served (permissive licence) · get_code("8720b5504bcf2aad")
How it ran

no invocation recorded.

Also present in other papers' code

Of 8 samples lifted from this paper's linked repositories, 8 carry a function identity and 0 do not. Of the 8 distinct functions with an identity, 3 are also present in other papers whose code we hold (0 with at least one paper dated earlier, 0 the same month, 3 later) and 5 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
broadcast_dims 1 other paper
1 later
2409.19433 (2024-09)
RMLR: Extending Multinomial Logistic Regression into General Geometries
ensure_sym 1 other paper
1 later
2409.19433 (2024-09)
RMLR: Extending Multinomial Logistic Regression into General Geometries
sum_bcastdims 1 other paper
1 later
2409.19433 (2024-09)
RMLR: Extending Multinomial Logistic Regression into General Geometries

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

Electroencephalography (EEG) provides access to neuronal dynamics noninvasively with millisecond resolution, rendering it a viable method in neuroscience and healthcare. However, its utility is limited as current EEG technology does not generalize well across domains (i.e., sessions and subjects) without expensive supervised re-calibration. Contemporary methods cast this transfer learning (TL) problem as a multi-source/-target unsupervised domain adaptation (UDA) problem and address it with deep learning or shallow, Riemannian geometry aware alignment methods. Both directions have, so far, failed to consistently close the performance gap to state-of-the-art domain-specific methods based on tangent space mapping (TSM) on the symmetric, positive definite (SPD) manifold. Here, we propose a machine learning framework that enables, for the first time, learning domain-invariant TSM models in an end-to-end fashion. To achieve this, we propose a new building block for geometric deep learning, which we denote SPD domain-specific momentum batch normalization (SPDDSMBN). A SPDDSMBN layer can transform domain-specific SPD inputs into domain-invariant SPD outputs, and can be readily applied to multi-source/-target and online UDA scenarios. In extensive experiments with 6 diverse EEG brain-computer interface (BCI) datasets, we obtain state-of-the-art performance in inter-session and -subject TL with a simple, intrinsically interpretable network architecture, which we denote TSMNet.

For agents

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

get_harvested_code_for_paper("2206.01323")
get_code_for_paper("2206.01323")
have("2206.01323")

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

curl https://syntology.ai/api/ran/2206.01323.json

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

[![Syntology run record](https://syntology.ai/api/ran/2206.01323.svg)](https://syntology.ai/paper/2206.01323)

Connect an agent — have() is free.