OSCR

Whole-to-parts causation mechanism.

Code ↔ Paper

1 match between paragraphs of the paper and lines of its authors' code, computed by the harvester (lexical-v1). Click a colored paragraph or line to see its counterpart.

The 1 match
  1. [1] § Proof of concept › Method › Training ↔ script/train_bisection_transInj.py, lines 31–89 · score 0.61 · RAdam, training, epochs, optimization, CUDA, batch

Paper

Loaded from Europe PMC by your browser, not stored by OSCR: doi.org · Europe PMC

The paper is loaded when this pane is shown.

The authors' code

Python · 113 lines · 3.9 KB · no license · 1 match

  1. import argparse
  2. import os
  3. import shutil
  4. from turtle import color
  5. from typing import Iterable, Optional, Union
  6. import torch
  7. import torch.nn as nn
  8. import torch.optim as optim
  9. from torch.utils.data import DataLoader
  10. from model.main_model import BisectionTrans
  11. from dataset.colorfont import ColorFontPairDataset, NPZImagesDataset
  12. import utils.json_model as jm
  13. import numpy as np
  14. import random
  15. from script.init_modelInj import (init_modelInj)
  16. import math
  18. def torch_fix_seed(seed: int):
  19. print("seed is " + str(seed))
  20. random.seed(seed)
  21. np.random.seed(seed)
  22. # Pytorch random
  23. torch.manual_seed(seed)
  24. torch.cuda.manual_seed(seed)
  25. torch.cuda.manual_seed_all(seed)
  26. torch.backends.cudnn.deterministic =True
  27. torch.backends.cudnn.benchmark = False
  28. torch.use_deterministic_algorithms = True
  31. def main(args):
  33. torch.set_num_threads(4)
  34. torch.autograd.set_detect_anomaly(True)
  35. if len(args.label_dim) == 1:
  36. args.label_dim = [args.label_dim[0], args.label_dim[0]]
  37. elif len(args.label_dim) > 2:
  38. raise ValueError
  40. print (args.label_dim)
  42. if args.seed is not None:
  43. torch_fix_seed(args.seed)
  45. model = init_modelInj(args.label_dim, p_ch=args.p_ch, n_ch=args.n_ch, orth=args.orth, add_ch=args.add_ch, gp_bias=args.gp_bias)
  46. model.log_config(args.logdir)
  48. model.Gn_config(orth=args.orth)
  50. # device
  51. if torch.cuda.is_available():
  52. if args.gpu < 0:
  53. # logger.info("CPU will be used.")
  54. device = torch.device("cpu")
  55. elif args.gpu >= torch.cuda.device_count():
  56. # logger.warn(f"Specified GPU ID {gpu_id} is invalid. "
  57. # "CPU will be used instead.")
  58. device = torch.device("cpu")
  59. else:
  60. device = torch.device("cuda", args.gpu)
  61. else:
  62. device = torch.device("cpu")
  63. # logger.warn("No CUDA device is available. CPU will be used.")
  64. model.to(device)
  66. optimizer = optim.RAdam(
  67. [
  68. {"params": model.Gp.Gp0.parameters()},
  69. {"params": model.Gp.Gp1.parameters()},
  70. {"params": filter(lambda p: p.requires_grad, model.Gn.parameters()) },
  71. ],
  72. args.learning_rate,
  73. )
  75. model.train_config(
  76. optimizer=optimizer,
  77. commute=args.commute,
  78. learning_rate = args.learning_rate,
  79. aux_rate =args.aux_rate,
  80. batch_size = args.batch_size,
  81. color_random=args.color_random,
  82. add_ch = args.add_ch
  83. )
  85. dataset = NPZImagesDataset("color_font_all.npz")
  86. loaders = (DataLoader(dataset, batch_size=args.batch_size, shuffle=True, drop_last=True),
  87. DataLoader(dataset, batch_size=args.batch_size, shuffle=True, drop_last=True))
  89. model.train(loaders, args.epoch)
  91. if __name__ == "__main__":
  92. parser = argparse.ArgumentParser()
  93. parser.add_argument("--gpu", "-g", type=int, default=-1)
  94. parser.add_argument("--logdir", "-l")
  95. parser.add_argument("--seed", "-sd", type=int)
  97. parser.add_argument("--label-dim", "-ldim",type=int, nargs="*", default=32)
  98. parser.add_argument("--p_ch", "-pch", type=int, default=64)
  99. parser.add_argument("--n_ch", "-nch", type=int, default=32)
  100. parser.add_argument("--add_ch", "-addc", action="store_true", default=False)
  102. # Important parameters
  103. parser.add_argument("--learning-rate", "-lr", type=float, default=1e-4)
  104. parser.add_argument("--batch-size", "-bs", type=int, default=128)
  105. parser.add_argument("--epoch", "-e", type=int, default=1000)
  106. parser.add_argument("--commute", "-cm", type=float, default=1.0)
  107. parser.add_argument("--aux_rate", "-ar", type=float, default=0.0)
  108. parser.add_argument("--orth", "-or", action="store_true", default=False)
  109. parser.add_argument("--gp_bias", "-gp_b", action="store_true", default=False)
  111. parser.add_argument("--color_random", "-cro", action="store_false", default=True)
  112. parser.add_argument("--shape_split", '-ss', action="store_true", default=False)
  113. main(parser.parse_args())

train_bisection_transInj.py at commit 2e6f0e1, no license · at the source

Overview

Authors: Yoshiyuki Ohmura1, Yasuo Kuniyoshi1
ORCID iDs: Yoshiyuki Ohmura
  1. Department of Mechano-Informatics, Graduate School of Information Science and Technology, The University of Tokyo, Tokyo, Japan
Institutions: The University of Tokyo (Japan)
Journal: Frontiers in psychology, volume 17, article 1654139
Dates: received 26 June 2025; accepted 21 April 2026; published online 12 May 2026
Type: Research article · Language: English
License: CC BY
Identifiers: DOI 10.3389/fpsyg.2026.1654139 · PMID 42205955 · PMCID PMC13202721 · OpenAlex W7160946548
Open access: gold, a free copy (OpenAlex)
Status: code verified
Categories: computational modeling (no new data) (modality), none (in silico) (organism)
Methods: Statistics, Complexity, Machine learning
Keywords: algebraic structure control, asymmetry between cause and effect, downward causation, inter-level causation, neural network model
Journal subjects: Hypothesis and Theory
Topic: Neural Networks and Applications (Artificial Intelligence, Computer Science), according to OpenAlex
Funding: Japan Society for the Promotion of Science (25H00448)
Citations: not cited yet (Europe PMC); 39 references in the paper

Abstract

How can the whole have a causal effect on its parts? This is considered impossible because, if the supervenient whole is completely determined by its parts, then the whole-to-parts causation would be redundant. However, the exclusion argument did not assume a hierarchy of multiple supervenient functions and the existence of an inter-level negative feedback control mechanism. Here, we propose that this mechanism enables a causal effect from the whole to its parts. Feedback control typically involves two mechanisms: observing and controlling of the feedback error. These mechanisms can be implemented at two different levels of the hierarchy. We assume that the macro-level consists of a set of mathematical functions that supervene on physical neural states. An algebraic structure of these functions describes a macro-level equation that determines the feedback error. This equation is independent of an external cause, introducing new causal power to the micro-level while avoiding overdetermination. Modifying the micro synaptic weights within the neural networks via inter-level negative feedback control is a whole-to-parts causation mechanism. It should be noted that this paper does not intend to take a position on the ontology surrounding downward causation.

Reproduced under the paper's license (CC BY), from the paper cited above.

Repository

Its files are read in the Code ↔ Paper reader above, with 1 match between paragraphs and lines of code.

Yoshiyuki-Ohmura/DownwardCausation

License: none: the authors keep all their rights
State: the link answers, verified on 28 September 2026
Evidence: files inventoried
Commit: 2e6f0e1d2c431804e6f51658a19df57b39f57658, 17 September 2024
Languages: Python (14), Shell (2)
Size: 30 files, 16 scripts
Software Heritage: not archived
Found in: the text, “Training”
Holds: README, environment (env.yml)
Not found: license file, CITATION.cff, tests, continuous integration, documentation
Tools: PyTorch (13 files), NumPy (6 files)
Availability: 1 check, the latest on 28 September 2026: the link answers
  • 28 September 2026: the link answers
17 files
At the source: github.com/Yoshiyuki-Ohmura/DownwardCausation

Tracing map

Proposed by the machine: these links were found in the paper and verified at the source, without human review. The map will receive a Zenodo DOI once one of the paper's authors has validated it with their ORCID.

What the map holds:

  • 1 repository of the authors' code, each at its verified commit, with its license and how the link was found in the paper;
  • 16 scripts, each with its path and the digest of its content;
  • 1 match between paragraphs of the paper and lines of the code (method lexical-v1);
  • neither the text of the paper nor the code itself.

Its JSON (tracing-map.json) is deposited on Zenodo with its DOI once the map is validated.

Data

No dataset and no data link were found in the paper.

Data availability statement

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

Reproduced under the paper's license (CC BY), from the paper cited above.

Versions

The history of this record: each version stored by the harvester or made by a correction of its authors or of the maintainers of its code, and what changed in its facts. The texts of the paper (its abstract, its availability statements) are not part of it; versions that changed only those are not listed.

Version 2, 28 September 2026

  • Funding: added Japan Society for the Promotion of Science: 25H00448

Version 1, 28 September 2026: the first record

Recorded: type, language, journal, volume, pages, dates, 2 authors, 5 keywords, 27 references.

Cite

This paper

Ohmura, Y., & Kuniyoshi, Y. (2026). Whole-to-parts causation mechanism. Frontiers in psychology, 17, 1654139. https://doi.org/10.3389/fpsyg.2026.1654139

BibTeX

@article{ohmura2026whole,
author = {Ohmura, Yoshiyuki and Kuniyoshi, Yasuo},
title = {{Whole-to-parts causation mechanism}},
journal = {Frontiers in psychology},
year = {2026},
month = may,
volume = {17},
pages = {1654139},
publisher = {Frontiers Media SA},
issn = {1664-1078},
doi = {10.3389/fpsyg.2026.1654139},
url = {https://doi.org/10.3389/fpsyg.2026.1654139},
pmid = {42205955},
pmcid = {PMC13202721}
}

RIS

TY - JOUR
AU - Ohmura, Yoshiyuki
AU - Kuniyoshi, Yasuo
TI - Whole-to-parts causation mechanism
T2 - Frontiers in psychology
J2 - Front Psychol
PY - 2026
DA - 2026/05/12
VL - 17
SP - 1654139
SN - 1664-1078
PB - Frontiers Media SA
DO - 10.3389/fpsyg.2026.1654139
UR - https://doi.org/10.3389/fpsyg.2026.1654139
LA - en
ER -

CSL-JSON

{
"id": "10.3389/fpsyg.2026.1654139",
"type": "article-journal",
"title": "Whole-to-parts causation mechanism",
"container-title": "Frontiers in psychology",
"author": [
{
"family": "Ohmura",
"given": "Yoshiyuki"
},
{
"family": "Kuniyoshi",
"given": "Yasuo"
}
],
"container-title-short": "Front Psychol",
"volume": "17",
"page": "1654139",
"DOI": "10.3389/fpsyg.2026.1654139",
"PMID": "42205955",
"PMCID": "PMC13202721",
"ISSN": "1664-1078",
"publisher": "Frontiers Media SA",
"URL": "https://doi.org/10.3389/fpsyg.2026.1654139",
"language": "en",
"issued": {
"date-parts": [
[
2026,
5,
12
]
]
}
}

The tracing map gets a citation of its own once an author has validated it and it has a DOI.

Similar papers

The papers with a page that share the most with this one: the tools found in their code, their categories, datasets, cited references and authors, the rarest counting most.

[1] doi:10.1093/nc/niag013 [code]
Intrinsic units: identifying a system's causal grain.
Journal: Neuroscience of consciousness
In common: NumPy, none (in silico), computational modeling (no new data), 2 references
[2] doi:10.1038/s41467-026-74460-8 [code]
Spike-based alignment learning solves the weight transport problem.
Journal: Nature communications
In common: PyTorch, NumPy, computational modeling (no new data), 1 reference
[3] doi:10.3389/fncir.2026.1783892 [code]
Representations of facial features and surface quality in monkey area TE compared to neural network models.
Journal: Frontiers in neural circuits
In common: PyTorch, NumPy, computational modeling (no new data), 1 reference
[4] doi: [code]
Going deeper with morphologically detailed neural networks by simulation-based gradient propagation
Journal: Frontiers in computational neuroscience
In common: NumPy, none (in silico), computational modeling (no new data), 1 reference
[5] doi:10.1038/s42256-026-01254-4 [code]
Neural sampling from cognitive maps enables goal-directed imagination and planning.
Journal: Nature machine intelligence
In common: PyTorch, NumPy, none (in silico), computational modeling (no new data)
[6] doi:10.1038/s41586-026-10670-w [code]
Zero-shot design of drug-binding proteins via neural iterative selection-expansion.
Journal: Nature
In common: PyTorch, NumPy, none (in silico), computational modeling (no new data)
[7] doi:10.3389/frai.2026.1785867 [code]
A generalized logistic-logit function and its application to multi-layer perceptron and neuron segmentation.
Journal: Frontiers in artificial intelligence
In common: PyTorch, NumPy, 1 reference
[8] doi:10.1523/eneuro.0370-25.2026 [code]
Pretraining for Large-Scale Functional Connectome Fingerprinting Supports Generalization and Transfer Learning in Functional Neuroimaging.
Journal: eNeuro
In common: PyTorch, NumPy, 1 reference
[9] doi:10.1162/imag.a.1309 [code]
Probing the content of semantic representations in body-selective regions.
Journal: Imaging neuroscience (Cambridge, Mass.)
In common: PyTorch, NumPy, 1 reference
[10] doi:10.1186/s12859-026-06490-4 [code]
Tissueformer: extending single-cell foundation models to predict population-level phenotypes.
Journal: BMC bioinformatics
In common: PyTorch, NumPy, 1 reference

Contribute

The authors of this paper can claim it, correct its record and validate its tracing map, and the maintainers of its code (its owner, or a public member of its organization) correct what it says of their repository; anyone signed in can ask for its removal. Every request goes to OSCR's own machine, which answers it; your account page follows them.

Sign in with ORCID to claim this paper as one of its authors, correct its record or validate its tracing map: when the paper's metadata lists your ORCID iD, you are recognized at once. Maintainers of its code: sign in with GitHub, then claim the repository on your account page.

Request its removal

To ask OSCR to remove this record, the copies of its authors' scripts or its tracing map, use the removal request page: signed in, you say who you are, what to remove and why, then review and confirm the request. Published rules decide every request (how).

Discussion, reproductions, activity

Discussion: questions and error reports about this paper and its code, from signed-in readers and its authors. It opens with sign-in.

Reproductions: reports from readers who ran the authors' code: what they reproduced, with which environment, commit and data. It opens with sign-in.

Activity: what happens around this paper: new versions of its record, its map's validation, discussions and reproductions. It opens with sign-in.