Feature Optimization for Time Series Forecasting via Novel Randomized Uphill Climbing
This work addresses the need for accurate and transparent forecasting tools for resource-constrained institutions, energy regulators, and climate risk NGOs, offering an incremental improvement by decoupling feature discovery from deep learning.
The paper tackles the problem of feature optimization for multivariate time series forecasting by generalizing Randomized Uphill Climbing into a model-agnostic framework, resulting in a method that promises faster iteration cycles, lower energy consumption, and greater interpretability compared to GPU-heavy deep learning approaches.
Randomized Uphill Climbing is a lightweight, stochastic search heuristic that has delivered state of the art equity alpha factors for quantitative hedge funds. I propose to generalize RUC into a model agnostic feature optimization framework for multivariate time series forecasting. The core idea is to synthesize candidate feature programs by randomly composing operators from a domain specific grammar, score candidates rapidly with inexpensive surrogate models on rolling windows, and filter instability via nested cross validation and information theoretic shrinkage. By decoupling feature discovery from GPU heavy deep learning, the method promises faster iteration cycles, lower energy consumption, and greater interpretability. Societal relevance: accurate, transparent forecasting tools empower resource constrained institutions, energy regulators, climate risk NGOs to make data driven decisions without proprietary black box models.