FlowState: Sampling-Rate-Equivariant Time-Series Forecasting
Researchers have introduced FlowState, a new time-series foundation model designed for enhanced adaptability and efficiency. Unlike previous transformer-based models, FlowState utilizes a state space model encoder paired with a functional basis decoder to achieve sampling-rate-equivariance. This architecture allows for continuous-time modeling and dynamic adjustment of forecasting horizons without retraining, enabling generalization across all temporal resolutions. Despite its smaller size, FlowState has demonstrated state-of-the-art performance on the GIFT-Eval benchmark and superior adaptability to unseen sampling rates. AI
IMPACT Introduces a novel architecture for time-series forecasting that generalizes across sampling rates, potentially improving efficiency and accuracy in applications like financial modeling and sensor data analysis.