Can I read Scaling Laws in Time Series Forecasting on EtoBox?

Scaling Laws in Time Series Forecasting by shiyuthu is a document available to read on EtoBox.

What is Scaling Laws in Time Series Forecasting about?

This document presents a comprehensive scaling law theory for time series forecasting, emphasizing the impact of look-back horizon on model performance. It empirically validates the existence of scaling behaviors related to dataset size and model complexity, while also identifying optimal horizons for different models. The findings aim to inspire future developments in forecasting models tailored for specific datasets and foundational models in time series forecasting.

Author

shiyuthu

Language

EN