ADAPTive Input Training for Many-to-One Pre-Training on Time-Series Classification
This addresses a fundamental problem in building foundation models for time-series data, enabling models to learn from diverse datasets, though it appears incremental as it builds on existing pre-training methods.
The paper tackled the challenge of many-to-one pre-training for time-series classification, where existing methods struggle to generalize when multiple datasets are added during pre-training, and introduced ADAPT, a new paradigm that efficiently aligns physical properties to enable mixed-batch training, achieving state-of-the-art performance on 162 datasets.
Recent work on time-series models has leveraged self-supervised training to learn meaningful features and patterns in order to improve performance on downstream tasks and generalize to unseen modalities. While these pretraining methods have shown great promise in one-to-many scenarios, where a model is pre-trained on one dataset and fine-tuned on a downstream dataset, they have struggled to generalize to new datasets when more datasets are added during pre-training. This is a fundamental challenge in building foundation models for time-series data, as it limits the ability to develop models that can learn from a large variety of diverse datasets available. To address this challenge, we present a new pre-training paradigm for time-series data called ADAPT, which can efficiently align the physical properties of data in the time-series domain, enabling mixed-batch pre-training despite the extreme discrepancies in the input sizes and channel dimensions of pre-training data. We trained on 162 time-series classification datasets and set new state-of-the-art performance for classification benchmarks. We successfully train a model within the time-series domain on a wide range of datasets simultaneously, which is a major building block for building generalist foundation models in time-series domains.