HP-GSBS: Accelerating greedy state boundary search with reusable interval statistics
Abstract
Greedy State Boundary Search (GSBS) identifies transitions between relatively stable neural activity patterns in continuous recordings. Repeated template reconstruction and correlation evaluation make analyses across recordings and parameter settings computationally expensive. We present HP-GSBS, an open-source implementation that expresses the matching objective as additive interval contributions and computes t-distance from reusable correlation statistics. Streaming candidate reduction, sequential local refinement, and active-block reuse complement CPU, PyTorch CUDA, and Triton execution paths. Across seven controlled workloads with two input realizations each, median same-input speedups range from 154 to 1,730. All 168 optimized executions reproduce complete reference boundary histories and selected state counts, with a maximum absolute t-distance difference of 1.88×10−11. Four publicly supplied BBBD EEG analysis matrices provide complementary cross-backend validation. A naturalistic fMRI application segments 4,230 angular-gyrus voxels in a group response and 16 participants; the 17 CPU fits total 213.10 seconds. Mean pairwise boundary alignment exceeds a circular-shift baseline, illustrating cohort-level use of the saved segmentations. Versioned results, recording-level parallelism, and optional low-memory execution support varied analysis scales. HP-GSBS connects an explicit mathematical reformulation of repeated computation to efficient and reproducible neural-state analysis.
Related
- Reusable modular architecture enables flexible cognitive operations in the mouse brain and artificial recurrent networks
- Reusable and single-use endoscopes: infection prevention and sustainability considerations - a narrative review.
- Xenograft supervision and joint methylation-fragment-end encoding define a reusable molecular scoring layer for cell-free tumor DNA
- A greedy nearest-neighbor approach to quantify site revisitation: comparing two sympatric raven species
- Increasing Width Allows Greedy Layer-wise Training to Rival End-to-End Backpropagation in Self-Supervised Learning
BibTeX
@misc{zhao2026hpgsbs,
title = {{HP-GSBS: Accelerating greedy state boundary search with reusable interval statistics}},
author = {Yanzhe Zhao and Yongzhang Ji},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.25898},
eprint = {2610.25898},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.25898},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - HP-GSBS: Accelerating greedy state boundary search with reusable interval statistics
AU - Yanzhe Zhao
AU - Yongzhang Ji
PY - 2026
DA - 2026/10/23
PB - arXiv
DO - 10.48550/arxiv.2610.25898
UR - https://arxiv.org/abs/2610.25898
AB - Greedy State Boundary Search (GSBS) identifies transitions between relatively stable neural activity patterns in continuous recordings. Repeated template reconstruction and correlation evaluation make analyses across recordings and parameter settings computationally expensive. We present HP-GSBS, an open-source implementation that expresses the matching objective as additive interval contributions and computes t-distance from reusable correlation statistics. Streaming candidate reduction, sequential local refinement, and active-block reuse complement CPU, PyTorch CUDA, and Triton execution paths. Across seven controlled workloads with two input realizations each, median same-input speedups range from 154 to 1,730. All 168 optimized executions reproduce complete reference boundary histories and selected state counts, with a maximum absolute t-distance difference of 1.88×10^−11. Four publicly supplied BBBD EEG analysis matrices provide complementary cross-backend validation. A naturalistic fMRI application segments 4,230 angular-gyrus voxels in a group response and 16 participants; the 17 CPU fits total 213.10 seconds. Mean pairwise boundary alignment exceeds a circular-shift baseline, illustrating cohort-level use of the saved segmentations. Versioned results, recording-level parallelism, and optional low-memory execution support varied analysis scales. HP-GSBS connects an explicit mathematical reformulation of repeated computation to efficient and reproducible neural-state analysis.
ER -