TimeBlocks: Foundational and Continual Time-Series Blockbase -- Extended Version
2026-06-01 • Machine Learning
Machine LearningDatabases
AI summaryⓘ
The authors notice that current big time-series models struggle with real-time data because they're too large and can't easily update themselves. They propose TimeBlocks, a method that builds smaller, flexible models from reusable pieces to better handle real-time streaming data. Their approach also includes StreamCore, which keeps a small but good summary of the data to help models update continuously. Tests on different tasks and data sets show that TimeBlocks creates models that work better than older methods. Overall, the authors aim to make time-series modeling more efficient and adaptable for live data streams.
time-series datafoundational modelsreal-time processingmodel calibrationmodular modelsdata streamsmodel routingcontinual learninglightweight modelsStreamCore
Authors
David Campos, Bin Yang, Tung Kieu, Lei Chen, Chenjuan Guo, Christian S. Jensen
Abstract
The ongoing digitization has led to a proliferation of time-series data streams that monitor a variety of processes, from which valuable insights may be obtained. Further, the emergence of successful foundational language models begs the question of whether it is possible to achieve time-series models with the foundational properties of handling multiple tasks, while being sufficiently lightweight to allow real-time data stream processing. Existing foundational time-series models are often large and only effective in offline settings without stringent time and computational constraints, and where repeated model calibration is not needed. However, when applied to data streams, these models are ineffective due to their size and lack of support for continual calibration, which compromise their ability to deliver accurate real-time responses, their durability, and their deployability in hardware-limited settings. We propose TimeBlocks to enable versatile time-series processing by facilitating the efficient building of lightweight models suitable for multiple tasks under variable conditions. In particular, the method maintains a pool of interchangeable and modular model blocks that can be used to construct new time-series models. When presented with specific time-series data, a routing strategy iteratively selects the most suitable blocks to construct a lightweight and accurate model for the data. We equip TimeBlocks with a method called StreamCore to build a representative small subset of the data stream, which preserves a guaranteed approximation of the stream over time, enabling continual model calibration. An experimental study on multiple data sets and covering multiple tasks shows that TimeBlocks enables to build models capable of outperforming existing baselines.