SEA-TS: Self-Evolving Agent for Autonomous Code Generation of Time Series Forecasting Algorithms
#SEA-TS #self-evolving agent #autonomous code generation #time series forecasting #algorithms #AI automation #machine learning
📌 Key Takeaways
- SEA-TS is a self-evolving agent designed for autonomous code generation of time series forecasting algorithms.
- It automates the development of forecasting models, reducing manual coding effort.
- The agent adapts and improves its code generation capabilities over time through self-evolution.
- This innovation aims to enhance efficiency and accuracy in time series analysis tasks.
📖 Full Retelling
arXiv:2603.04873v1 Announce Type: new
Abstract: Accurate time series forecasting underpins decision-making across domains, yet conventional ML development suffers from data scarcity in new deployments, poor adaptability under distribution shift, and diminishing returns from manual iteration. We propose Self-Evolving Agent for Time Series Algorithms (SEA-TS), a framework that autonomously generates, validates, and optimizes forecasting code via an iterative self-evolution loop. Our framework int
🏷️ Themes
AI Automation, Time Series Forecasting
Entity Intersection Graph
No entity connections available yet for this article.
Original Source
--> Computer Science > Artificial Intelligence arXiv:2603.04873 [Submitted on 5 Mar 2026] Title: SEA-TS: Self-Evolving Agent for Autonomous Code Generation of Time Series Forecasting Algorithms Authors: Longkun Xu , Xiaochun Zhang , Qiantu Tuo , Rui Li View a PDF of the paper titled SEA-TS: Self-Evolving Agent for Autonomous Code Generation of Time Series Forecasting Algorithms, by Longkun Xu and 3 other authors View PDF HTML Abstract: Accurate time series forecasting underpins decision-making across domains, yet conventional ML development suffers from data scarcity in new deployments, poor adaptability under distribution shift, and diminishing returns from manual iteration. We propose Self-Evolving Agent for Time Series Algorithms (SEA-TS), a framework that autonomously generates, validates, and optimizes forecasting code via an iterative self-evolution loop. Our framework introduces three key innovations: (1) Metric-Advantage Monte Carlo Tree Search (MA-MCTS), which replaces fixed rewards with a normalized advantage score for discriminative search guidance; (2) Code Review with running prompt refinement, where each executed solution undergoes automated review followed by prompt updates that encode corrective patterns, preventing recurrence of similar errors 3) Global Steerable Reasoning, which compares each node against global best and worst solutions, enabling cross-trajectory knowledge transfer. We adopt a MAP-Elites archive for architectural diversity. On the public Solar-Energy benchmark, SEA-TS generated code achieves a 40% MAE reduction relative to TimeMixer, surpassing state-of-the-art methods. On proprietary datasets, SEA-TS generated code reduces WAPE by 8.6% on solar PV forecasting and 7.7% on residential load forecasting compared to human-engineered baselines, and achieves 26.17% MAPE on load forecasting versus 29.34% by TimeMixer. Notably, the evolved models discover novel architectural patterns--including physics-informed monotonic decay heads encodi...
Read full article at source