Synapse: Adaptive Arbitration of Complementary Expertise in Time Series Foundational Models

TL;DR

This paper introduces Synapse, an adaptive arbitration framework that dynamically combines multiple pre-trained Time Series Foundational Models to improve forecasting accuracy across diverse tasks and settings.

Contribution

The paper presents a novel arbitration method for TSFMs that adjusts model weights based on context-dependent performance, enhancing forecasting accuracy over existing ensembling techniques.

Findings

Synapse outperforms popular ensembling methods and individual TSFMs.

Adaptive weighting improves forecast accuracy across various domains.

The framework effectively leverages complementary expertise of different TSFMs.

Abstract

Pre-trained Time Series Foundational Models (TSFMs) represent a significant advance, capable of forecasting diverse time series with complex characteristics, including varied seasonalities, trends, and long-range dependencies. Despite their primary goal of universal time series forecasting, their efficacy is far from uniform; divergent training protocols and data sources cause individual TSFMs to exhibit highly variable performance across different forecasting tasks, domains, and horizons. Leveraging this complementary expertise by arbitrating existing TSFM outputs presents a compelling strategy, yet this remains a largely unexplored area of research. In this paper, we conduct a thorough examination of how different TSFMs exhibit specialized performance profiles across various forecasting settings, and how we can effectively leverage this behavior in arbitration between different time series models. We specifically analyze how factors such as model selection and forecast horizon distribution can influence the efficacy of arbitration strategies. Based on this analysis, we propose Synapse, a novel arbitration framework for TSFMs. Synapse is designed to dynamically leverage a pool of TSFMs, assign and adjust predictive weights based on their relative, context-dependent performance, and construct a robust forecast distribution by adaptively sampling from the output quantiles of constituent models. Experimental results demonstrate that Synapse consistently outperforms other popular ensembling techniques as well as individual TSFMs, demonstrating Synapse's efficacy in time series forecasting.