Python in Science Conference (SciPy)
Published July 10, 2024 | Version v1
Poster Open

Employing the strengths of Generative AI supports the execution of time series analysis and forecasting

Authors/Creators

  • 1. Georgia Institute of Technology, College of Computing

Contributors

  • 1. GDI
  • 2. SLB
  • 3. University of North Carolina
  • 4. Curvenote
  • 5. Deloitte
  • 6. Aptos
  • 7. Arm

Description

This poster explores the use of Generative AI models for time series analysis and forecasting, specifically in the context of energy consumption. It compares traditional statistical methods, such as ARIMA, with advanced AI-based techniques like AutoGluon-TimeSeries, xLSTM, and TimeGPT. The study aims to demonstrate the efficiency and accuracy of these methods using real-world energy data from the PJM Interconnection LLC. Results indicate that AI-based models, particularly xLSTM and AutoGluon-TimeSeries, outperform traditional models in forecasting accuracy, showcasing their potential for better resource management and decision-making in climate change mitigation.

Files

YJC_poster.pdf

Files (2.9 MB)

Name Size Download all
md5:00b0741f1be22c8a56c5052d6ce379a7
2.9 MB Preview Download