ELATE: Evolutionary Language model for Automated Time-series Engineering
This addresses the need for efficient and domain-aware automation in time-series analysis, though it appears incremental as it builds on existing evolutionary and language model approaches.
The paper tackles the problem of automating feature engineering for time-series prediction, which is often manual and computationally costly, by introducing ELATE, an evolutionary framework with a language model that improves forecasting accuracy by an average of 8.4% across domains.
Time-series prediction involves forecasting future values using machine learning models. Feature engineering, whereby existing features are transformed to make new ones, is critical for enhancing model performance, but is often manual and time-intensive. Existing automation attempts rely on exhaustive enumeration, which can be computationally costly and lacks domain-specific insights. We introduce ELATE (Evolutionary Language model for Automated Time-series Engineering), which leverages a language model within an evolutionary framework to automate feature engineering for time-series data. ELATE employs time-series statistical measures and feature importance metrics to guide and prune features, while the language model proposes new, contextually relevant feature transformations. Our experiments demonstrate that ELATE improves forecasting accuracy by an average of 8.4% across various domains.