SiBBlInGS: Similarity-driven Building-Block Inference using Graphs across States

TL;DR

SiBBlInGS is a graph-based framework that infers interpretable building blocks in multi-state time series data, capturing inter- and intra-state variability, and handling missing data, demonstrated on synthetic and real-world datasets.

Contribution

The paper introduces SiBBlInGS, a novel graph-based dictionary learning method that uncovers sparse, interpretable building blocks across states, accounting for variability and missing data.

Findings

Successfully reveals complex patterns in synthetic and real data.

Robust to noise and missing samples in diverse datasets.

Identifies state-specific and invariant components effectively.

Abstract

Time series data across scientific domains are often collected under distinct states (e.g., tasks), wherein latent processes (e.g., biological factors) create complex inter- and intra-state variability. A key approach to capture this complexity is to uncover fundamental interpretable units within the data, Building Blocks (BBs), which modulate their activity and adjust their structure across observations. Existing methods for identifying BBs in multi-way data often overlook inter- vs. intra-state variability, produce uninterpretable components, or do not align with properties of real-world data, such as missing samples and sessions of different duration. Here, we present a framework for Similarity-driven Building Block Inference using Graphs across States (SiBBlInGS). SiBBlInGS offers a graph-based dictionary learning approach for discovering sparse BBs along with their temporal traces, based on co-activity patterns and inter- vs. intra-state relationships. Moreover, SiBBlInGS captures per-trial temporal variability and controlled cross-state structural BB adaptations, identifies state-specific vs. state-invariant components, and accommodates variability in the number and duration of observed sessions across states. We demonstrate SiBBlInGS's ability to reveal insights into complex phenomena as well as its robustness to noise and missing samples through several synthetic and real-world examples, including web search and neural data.