Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
Wavelet Canonical Coherence for Nonstationary Signals
Authors: Haibo Wu, Marina Knight, Keiland Cooper, Norbert Fortin, Hernando Ombao
NeurIPS 2025 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Through extensive simulation studies, we demonstrate that Wave Can Coh accurately recovers true coherence structures under both locally stationary and general nonstationary conditions. Application to local field potential (LFP) activity data recorded from the hippocampus reveals distinct dynamic coherence patterns between correct and incorrect memory-guided decisions, illustrating the capacity of the method to detect behaviorally relevant neural coordination. |
| Researcher Affiliation | Academia | 1Statistics Program, King Abdullah University of Science and Technology 2Department of Mathematics, The University of York 3Department of Neurobiology and Behavior, University of California, Irvine 4Center for the Neurobiology of Learning and Memory, University of California, Irvine |
| Pseudocode | Yes | Algorithm 1 Proposed Wave Can Coh estimation algorithm for nonstationary time series |
| Open Source Code | Yes | Code for implementing Wave Can Coh is available at https://github.com/mhaibo/Wave Can Coh.git. |
| Open Datasets | No | The real dataset used is not publicly available. |
| Dataset Splits | No | We simulate 500 independent replicates of two multivariate processes, {Xt} R4 and {Yt} R3, for t = 1, . . . , T, with T = 1024 and sampling rate fs = 100Hz. Specifically, we analyze LFP data from the rat Mitt, which included 40 correct-response trials and 32 incorrect-response trials. |
| Hardware Specification | Yes | in the experiments with T = 1024, a single replicate is completed within 2.5 seconds on a standard personal computer (Apple Mac, 16GB RAM, 6-core CPU) without resorting to parallel computing or to the use of a cluster, and all results reported in the paper can be obtained within 3 hours. |
| Software Dependencies | No | The process is constructed using non-decimated Haar wavelets... We employ Wave Can Coh framework with Haar wavelets... |
| Experiment Setup | Yes | In this section, we implement the proposed framework using simulated data under two distinct scenarios, one adhering to the Mv LSW assumptions underpinning our method, while the other introduces nonstationarity without strictly satisfying the Mv LSW assumptions. To demonstrate the practical utility of our proposed Wave Can Coh framework, we analyze LFP activity recorded from the hippocampus of rats engaged in a sequence memory task [2, 22]. The data were recorded using a 22-electrode microdrive implanted in the CA1 subregion to capture high-resolution LFP signals across all channels at a sampling rate of 1000Hz. We simulate 500 independent replicates of two multivariate processes, {Xt} R4 and {Yt} R3, for t = 1, . . . , T, with T = 1024 and sampling rate fs = 100Hz. |