HSF-India Trainee Fellow: Nikhil Jangid
Fellowship dates: Jun – Aug, 2024
Home Institution: Vellore Institute of Technology, India
Project: Anomaly Detection using XLSTM-VAE for Data Quality Monitoring in Liquid Argon Calorimeters
● A fully documented XLSTM-VAE model integrated with the IRIS-HEP anomaly detection infrastructure. ● Complete source code for training, evaluation, and inference to ensure the work can be reproduced and built upon. ● A trained XLSTM-VAE model capable of capturing longer temporal dependencies, leading to better anomaly detection in the detector data. ● Performance comparisons showing how this model stands against traditional LSTM-VAE models, especially in handling data drift over time. ● A comprehensive technical report detailing the methods, implementation, experiments, and results, ready for publication on platforms like arXiv or Zenodo.More information: My project proposal
Mentors:
-
Pratik Jawahar (CERN)
Current Status
Contact me: