Summary of Enabling Time-series Foundation Model For Building Energy Forecasting Via Contrastive Curriculum Learning, by Rui Liang et al.
Enabling Time-series Foundation Model for Building Energy Forecasting via Contrastive Curriculum Learning
by Rui Liang, Yang Deng, Donghua Xie, Fang He, Dan Wang
First submitted to arxiv on: 23 Dec 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper explores the application of foundation models (FMs) in building energy forecasting (BEF), a task that has seen limited success with conventional machine learning approaches. The authors highlight the shortcomings of fine-tuning FMs for BEF, citing both model and data perspectives. To address these limitations, they propose a novel training method based on contrastive curriculum learning. This approach optimizes the ordering of training data in the context of time-series FM adaptation. Experiment results demonstrate that this method can improve zero/few-shot performance by 14.6% compared to existing FMs. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper is about using special kinds of computer models called foundation models to predict energy usage in buildings. These models are normally used for other tasks, but they don’t work very well when trying to forecast building energy usage. The researchers show that there are problems with both the models and the data when trying to adapt them for this task. To fix these issues, they developed a new way of training the models using something called contrastive curriculum learning. This method helps the model learn from the training data in the best possible way for building energy forecasting. The results show that this new approach can make big improvements over previous methods. |
Keywords
» Artificial intelligence » Curriculum learning » Few shot » Fine tuning » Machine learning » Time series
