Authors: Mohamed Mejri, Chandramouli Amarnath, Abhijit Chatterjee
Published on: February 03, 2024
Impact Score: 8.15
Arxiv code: Arxiv:2402.01999
Summary
- What is new: Reframing the online nonlinear time-series forecasting problem as a linear hyperdimensional forecasting issue.
- Why this is important: Offline deep forecasting models can’t adapt to changes effectively, and online models are often too expensive and complex.
- What the research proposes: A new framework, TSF-HD, that uses a novel co-training framework for mapping nonlinear time-series data to a high-dimensional space for efficient forecasting.
- Results: TSF-HD outperforms the state of the art in delivering faster and more efficient forecasting for both short-term and long-term projections.
Technical Details
Technological frameworks used: TSF-HD
Models used: Co-training framework for hyperdimensional mapping and linear hyperdimensional predictor.
Data used: nan
Potential Impact
Companies in stock market forecasting, weather forecasting, and demand forecasting may benefit from or be disrupted by TSF-HD.
Want to implement this idea in a business?
We have generated a startup concept here: HyperForecast.
Leave a Reply