Sidharth SS

Computational Physics & Machine Learning Researcher

I am an integrated BS-MS physics student at the Indian Institute of Science Education and Research (IISER) Mohali, specialising in experimental particle physics and focusing on the CMS experiment at CERN. My strong interest in mathematics naturally led me to deep learning, which I have explored for over four years and now apply in my research. I enjoy combining physics and AI to tackle challenging scientific problems and quickly adapt to new concepts to drive innovative projects. Beyond research, I write about AI in my blog, sharing how it works and how it impacts science. I also contribute to open-source projects, helping build and improve tools for machine learning and scientific computing.

Work Experience

Master's Thesis

TIFR Mumbai & IISER Mohali | Feb 2025 – Present

Pursuing Master's thesis under Dr. Shashi Dugad, focusing on the reconstruction of hadronic and electromagnetic showers in the High Granularity Calorimeter (HGCAL) of the CMS detector. The work involves developing a graph neural network (GNN)-based pipeline to predict particle energy and enhance the identification of shower topologies in the highly segmented calorimeter environment.

Research Intern - CMS Experiment

University of Alabama | Feb 2024–Aug 2024

Worked on the project “Self-Supervised Learning for End-to-End Particle Reconstruction for the CMS Experiment” under Dr. Sergei Gleyzer. Focused on using self-supervised machine learning methods, improving how particles are identified and how collision events are reconstructed. Developed approaches to handle complex detector data while making the models more accurate and reliable.

Research Intern - Climate Prediction

IISER Mohali | Apr 2024–Sep 2024

As a side project under Dr. Raju Attada at IISER Mohali, I developed advanced deep learning models using 70 years of time-series and multidimensional climate data from ERA5 to predict heavy anomalous rainfall events in the Uttarakhand region. The work focused on using long-term climate trends and spatial patterns to improve forecast accuracy, helping to detect extreme rainfall early and spot unusual weather events.

Research Intern - X-ray Astrophysics

IISER Mohali | Nov 2023–Mar 2024

Under Dr. Aru Beri's guidance, worked on neutron stars and black holes in X-ray binaries by extracting light curves from observational data and using physical analysis to study their behaviour and determine the nature of the compact objects

Research Intern, Associate Member - SWGO Experiment

University of Padova & INFN Italy | Feb 2023–Sep 2023

Worked on developing a well-defined parametrisation model for the lateral distribution of secondary particles in extensive air showers to optimise the Southern Wide-field Gamma-ray Observatory (SWGO) detector array layout. Used Monte Carlo simulations (CORSIKA) to study electron, photon, and muon density patterns across different energies and angles. Developed a machine learning tool (CurvPy) to extract key parameters, allowing quick and accurate flux predictions without rerunning heavy simulations.

Research Intern – Protein Structure Analysis with Deep Learning

IISER Mohali | Dec 2022 – Feb 2023

Worked under Dr. Shashi Bhushan Pandit on analysing protein structures and applying deep learning techniques to study structural properties and functional patterns.

Publications

Chebyshev Polynomial-Based Kolmogorov-Arnold Networks: An Efficient Architecture for Nonlinear Function Approximation
Sidharth, S. S., & Gokul, R. (2024)
End-To-End Optimization of the Layout of a Gamma Ray Observatory
Dorigo, T., Aehle, M., Donini, J., Doro, M., Gauger, N. R., Izbicki, R., Lee, A., Masserano, L., Nardi, F., Sidharth, S. S., & Shen, A. (2023)
Acknowledgement: The expected potential of hadronic PeVatron searches with spectral γ-ray data from the Southern Wide-field Gamma-ray Observatory
Angüner, E. O., Ergin, T. (2023)

Upcoming Publications

Graph Neural Networks for Particle Reconstruction in High Granularity Calorimeter (HGCAL)
Sidharth, S. S., Dugad, S., et al. (In Preparation)

Projects

Deep-KAN

Open Source

Developed a Python Library named Deep-KAN for Kolmogorov Arnold Networks (KAN), facilitating efficient implementation and deployment for various machine learning tasks.

References

Dr. Shashi Dugad

Professor, Tata Institute of Fundamental Research (TIFR) Mumbai

Dr. Tommaso Dorigo

Senior Researcher, INFN Padova & CMS Collaboration, CERN

Professional Links

Technical Skills

  • Python
  • C++
  • Bash
  • PyTorch
  • TensorFlow
  • ROOT
  • GEANT4
  • HTML

Education

  • Integrated BS-MS in Physics
    IISER Mohali
    2021 – Present
  • Higher Secondary Certificate
    Board of Secondary Education Kerala
    2020 (95%)
  • Secondary School Certificate
    Board of Secondary Education Kerala
    2018 (99%)

Conferences

  • European Astronomical Society Annual Meeting
    July 2024
  • 16th International Workshop on Heavy Quarkonium
    Feb-Mar 2024
  • 8th General Meeting of the SWGO Collaboration
    Apr 2023
  • 25th DAE-BRNS HEP Symposium
    Dec 2022

Metrics & Impact

133
Citations
2
H-Index
1
i10-Index
92
GitHub Stars