Tree-based Learning for High-Fidelity Prediction of Chaos
This work addresses the problem of chaotic system prediction for researchers and practitioners, offering a more accessible method, though it appears incremental as it builds on existing tree-based and embedding techniques.
The authors tackled the challenge of model-free forecasting for chaotic systems by introducing TreeDOX, a tree-based approach that eliminates hyperparameter tuning, achieving state-of-the-art performance on benchmarks like the Henon map, Lorenz system, and real-world Southern Oscillation Index.
Model-free forecasting of the temporal evolution of chaotic systems is crucial but challenging. Existing solutions require hyperparameter tuning, significantly hindering their wider adoption. In this work, we introduce a tree-based approach not requiring hyperparameter tuning: TreeDOX. It uses time delay overembedding as explicit short-term memory and Extra-Trees Regressors to perform feature reduction and forecasting. We demonstrate the state-of-the-art performance of TreeDOX using the Henon map, Lorenz and Kuramoto-Sivashinsky systems, and the real-world Southern Oscillation Index.