Self-Supervised Learning via Flow-Guided Neural Operator on Time-Series Data

TL;DR

This paper introduces FGNO, a flow-guided neural operator framework for self-supervised learning on time-series data, which improves representation flexibility and accuracy across biomedical applications by leveraging noise as a learning signal.

Contribution

The paper proposes a novel flow-guided neural operator framework that treats corruption level as a learnable factor, enhancing SSL for time-series data with multi-resolution feature extraction.

Findings

Achieves up to 35% AUROC improvement in neural decoding

Reduces RMSE by 16% in skin temperature prediction

Improves accuracy and macro-F1 on SleepEDF in low-data settings

Abstract

Self-supervised learning (SSL) is a powerful paradigm for learning from unlabeled time-series data. However, popular methods such as masked autoencoders (MAEs) rely on reconstructing inputs from a fixed, predetermined masking ratio. Instead of this static design, we propose treating the corruption level as a new degree of freedom for representation learning, enhancing flexibility and performance. To achieve this, we introduce the Flow-Guided Neural Operator (FGNO), a novel framework combining operator learning with flow matching for SSL training. FGNO learns mappings in functional spaces by using Short-Time Fourier Transform to unify different time resolutions. We extract a rich hierarchy of features by tapping into different network layers and flow times that apply varying strengths of noise to the input data. This enables the extraction of versatile representations, from low-level patterns to high-level global features, using a single model adaptable to specific tasks. Unlike prior generative SSL methods that use noisy inputs during inference, we propose using clean inputs for representation extraction while learning representations with noise; this eliminates randomness and boosts accuracy. We evaluate FGNO across three biomedical domains, where it consistently outperforms established baselines. Our method yields up to 35% AUROC gains in neural signal decoding (BrainTreeBank), 16% RMSE reductions in skin temperature prediction (DREAMT), and over 20% improvement in accuracy and macro-F1 on SleepEDF under low-data regimes. These results highlight FGNO's robustness to data scarcity and its superior capacity to learn expressive representations for diverse time series.