References

CheungOCD

N. Cheung, Structural synaptogenesis superior to functional modulation in a pruning-based recurrent network model of OCD. Frontiers in Computational Neuroscience 20 (2026), 1799705. Published 20 July 2026. https://doi.org/10.3389/fncom.2026.1799705. Computational motivation. The article's operational and biological qualifications are retained; no clinical hierarchy is imported.

CheungPruning

N. Cheung, Phased pruning in neural networks recapitulates selectivity–fragility trade-offs in brain development. Scientific Reports (2026). Published online 13 July 2026. https://doi.org/10.1038/s41598-026-62244-5. A different computational task and architecture.

CheungModes

N. Cheung, Divergent mechanisms of antidepressant efficacy: a unified computational comparison of synaptogenesis, stabilization, and tonic inhibition in a model of depression. Cureus 18, no. 3 (2026), e105040. Published 11 March 2026. https://doi.org/10.7759/cureus.105040. Intervention motivation, not a clinical premise.

CheungPublicCode

N. Cheung, 069av2.py. Public file in Structural-Synaptogenesis-Superior-to-Functional-Modulation-in-a-Network-Model-of-OCD. https://github.com/cheungngo/Structural-Synaptogenesis-Superior-to-Functional-Modulation-in-a-Network-Model-of-OCD/blob/main/069av2.py. Consulted for the task mappings and generator. Mutable branch record; no byte identity with an unprovided attachment, source checksum, or verified numerical reproduction is asserted.

DevelopmentBrief

N. Cheung, Psy03-D development brief: hidden-rule observability, switching metrics, masked repair, and certified matching. Supplied documentary text, undated. Its implementation descriptions are source-pattern records, not a complete versioned code deposit.

CheungPsy01A

N. Cheung, Psy01-A: Phased magnitude pruning in task-gated ReLU networks: cue certificates, stability, and conditional selectivity–fragility trade-offs. Undated manuscript, Psy01-A.tex, supplied in full. Relevant source records: thm:cueobstruction, thm:nesting, thm:accuracystability, and prop:harddiscontinuity. No architecture identity is presumed.

CheungPsy02A

N. Cheung, Psy02-A: Conditional scalar-carrier bounds and operational separation: a mathematical framework for evaluating Orch-OR hypotheses. Undated manuscript, Psy02-A.tex, supplied in full. Relevant records: prop:instrument, thm:nonidentifiable, and thm:nonentailment. Its scalar and reduction models are not imported.

CheungPsy02B

N. Cheung, Psy02-B: Quantitative viability and identifiability of a branch-matched reduction pilot: complete records, collective mass energies, and a self-contained interface to the inherited analytic problem. Undated manuscript, Psy02-B-v2.tex, version 2, supplied in full. Relevant records: thm:process, thm:overlap, thm:sampling, and prop:nontransfer. The analytic continuation and admission obligations remain separate.

CheungPsy03A

N. Cheung, Psy03-A: Structural repair, noise adaptation, and dynamical modulation in pruned networks: intervention budgets, classical certificates, and a separately defined quantum extension. Undated manuscript, Psy03-A.tex, supplied in full. Relevant records: def:budget, def:matching, thm:descent, thm:ranking, thm:evaluation, and prop:rng.

CheungPsy03B

N. Cheung, Psy03-B: Irreversible structural memory and threshold-triggered feedback in pruned networks: longitudinal certificates and a quantum–classical extension. Undated manuscript, Psy03-B-v1.tex, version 1, supplied in full. Relevant records: cor:capacity, thm:conditional, prop:fixeddesign, thm:paired, and thm:operationalmemory. No permanent scar register is inferred for the modular source.

CheungPsy03C

N. Cheung, Psy03-C: Recurrent stability and probe identifiability in scar-constrained networks: task geometry, temporal certificates, and a separate quantum extension. Undated manuscript, Psy03-C-v1.tex, version 1, supplied in full. Relevant records: prop:repetition, thm:bayes, cor:battery, prop:identifiability, prop:register, and thm:qpilot. Its repeated-point task and recurrent state interface are not identified with the present task.

Hoeffding

W. Hoeffding, Probability inequalities for sums of bounded random variables. Journal of the American Statistical Association 58, no. 301 (1963), 13–30. https://doi.org/10.1080/01621459.1963.10500830. Here applied to independent or conditionally fresh sequence blocks.

EvciRigL

U. Evci, T. Gale, J. Menick, P. S. Castro, and E. Elsen, Rigging the lottery: making all tickets winners. In Proceedings of the 37th International Conference on Machine Learning, Proceedings of Machine Learning Research 119 (2020), 2943–2952. https://proceedings.mlr.press/v119/evci20a.html. A topology-update precedent, not an identity with the reported repair algorithm.

Watrous

J. Watrous, The Theory of Quantum Information. Cambridge University Press, 2018. Author's manuscript: https://cs.uwaterloo.ca/ watrous/TQI/. Cited for channels, instruments, trace distance, and binary state discrimination.

PyTorchGRU

PyTorch Contributors, GRU. Official PyTorch main documentation. https://docs.pytorch.org/docs/main/generated/torch.nn.GRU.html. Cited for gate conventions, initial-state and sequence interfaces, and interlayer dropout placement. An executed reproduction must pin its installed implementation.

PyTorchPruning

M. Paganini and PyTorch Contributors, Pruning Tutorial. Official PyTorch Tutorials. Created 22 July 2019; updated 2 November 2023; last-verified date stated as 5 November 2024. https://docs.pytorch.org/tutorials/intermediate/pruning_tutorial.html. Cited for original parameters, mask buffers, effective tensors, iterative masks, serialization, and removal of reparameterization. Its use of “permanent” for retaining current pruned values is not identified with a future regrowth-prohibition rule.

PyTorchRandom

PyTorch Contributors, Reproducibility. Official PyTorch documentation. https://docs.pytorch.org/docs/main/notes/randomness.html. Environment- and version-specific implementation guidance, not a proof of statistical independence.