Summary of Enhancing Indoor Temperature Forecasting Through Synthetic Data in Low-data Environments, by Zachari Thiry et al.
Enhancing Indoor Temperature Forecasting through Synthetic Data in Low-Data Environments
by Zachari Thiry, Massimiliano Ruocco, Alessandro Nocente, Michail Spitieris
First submitted to arxiv on: 7 Jun 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 The paper investigates the efficacy of data augmentation techniques using SoTA AI-based methods to generate synthetic data for indoor temperature forecasting, addressing the challenge of limited available data. The authors explore fusion strategies of real and synthetic data to improve forecasting models, alleviating the need for extensive time series data acquisition. They evaluate the performance of synthetic data generators and assess the utility of incorporating synthetically augmented data in a subsequent forecasting task using a simple model in two scenarios: expanding the training dataset and tackling dataset imbalances. The results highlight the potential of synthetic data augmentation in enhancing forecasting accuracy while mitigating training variance. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper is about making better predictions for indoor temperatures, which helps control heating and cooling systems more efficiently. There’s not enough data to make good predictions because most of it is collected when things are normal, but extreme temperature changes are hard to predict. To fix this, the authors look at ways to create fake data using AI-based methods that mimic real-world conditions. They test different approaches and find that combining real and fake data improves prediction accuracy. This could be useful for building management and HVAC systems. |
Keywords
* Artificial intelligence * Data augmentation * Synthetic data * Temperature * Time series
