Shashank Kumar Roy
A pdf version is available here: click to open or directly download
Current Affiliation
Institution International Centre for Theoretical Sciences - TIFR, Bangalore, India
Position PhD Research Scholar, Data Assimilation (Applied Mathematics)
e-mail shashank.roy@icts.res.in, shashankroy1997@gmail.com
linkedin linkedin.com/in/shashankroy/
Office address G-106, academic block, ICTS, Survey No. 151, Shivakote, Hesaraghatta Hobli,
Bangalore - 560 089, India.
Education
Doctor of Philosophy in Physics | Ongoing | July 2020-Present
International Centre for Theoretical Sciences, TIFR Bangalore, India
Masters in Physics | CPI 70.8 | 2017-2019
International Centre for Theoretical Sciences, TIFR Bangalore, India
Bachelors in Physics ( Honors ) | 84 % | 2014-2017
University of Delhi, Ramjas College, New Delhi, India
Intermediate in Science ( CISCE ) | 95.6% |2012-2014
Denobili School Mugma, Mugma, Dhanbad, Jharkhand, India
Publications
Shashank Kumar Roy, Amit Apte. Reconstructing Covariant Lyapunov Vectors from partial and noisy observations using Data Assimilation [Manuscript in Preparation]
Mandal P, Roy SK, Apte A. Probing robustness of nonlinear filter stability numerically using Sinkhorn divergence. Submitted to Physica D, arxiv link-https://arxiv.org/abs/2208.10810
Mandal P, Roy SK, Apte A. Stability of nonlinear filters-numerical explorations of particle and ensemble Kalman filters. In: 2021 Seventh Indian Control Conference (ICC) IEEE; 2021. p. 307–312, https://doi.org/10.1109/ICC54714.2021.9703185
Research Experience
Winter Project | International Centre for Theoretical Sciences | October 2019- Dec 2019 Sequential estimation of conditional distribution of the state of a 40-dimensional nonlinear dynamical system with partial obsevrations via Kalman Filtering.
Hackathon | A GAN model for distribution of Sea Surface Temperature| Oct – Dec 2022 Ecole Polytechnique, BNP Paribas, Fondation de l’Ecole polytechnique and Mercator Ocean Objective: Modeling distribution of the sea surface temperaure at 6 different locations and interpolate to a new location using a GAN to perform regression.
Hackathon | LSTM model to predict spatial time series for climate model emulation |7- 14 Sep 2021
3rd NOAA AI Workshop - Climate Informatics Joint Hackathon
Objective: Predict annual mean global distributions of temperature and precipitation given emissions and concentrations of key anthropogenic climate forcing: SO2, BC, CH4, and CO.
Summer Project | International Centre for Theoretical Sciences | June 2018 -Aug 2018 Analytical calculation of Stokes’ flow inside a Sphere via Vector Spherical Harmonics. generated by given surface flow. A python package mayavi was used to create 3D visualizations of the flow.
Resreach Internship | University of Delhi Innovation Project | Oct 2015 - Nov 2016 Project titled" An Interdisciplinary Study of Light Pollution in Indian Context (Extension)", was collection, analysis and quantification of light pollution from google light pollution map.
Achievements
- Secured 10th best score in IBM Quantum Challenge 2021
- Department of Atomic Energy Fellowship for pursuing PhD in Physics
- Joint Entrance Screening Test 2017, AIR-95, Percentile-98.8 and IIT-JAM 2017, AIR-259.
- Awarded ISC-2014 Science School Topper and 3rd topper at district level.
Teaching Experience
Teaching Assistant, Course on Time Series analysis, Data Science Department IISER Pune, India | Jan-April 2023| Instructor: Dr. Tulsi Ram Reddy view
Curating Jupyter notebooks for demonstrations and taking tutorial sessions.
Freelance | Advanced Physics Subject Matter Expert | Chegg India | July 2020- July 2021 Solving physics problems at college and university level on an online platform.
Skills and Certifications
Scientific : Bayesian Data Assimilation | Ensemble Kalman Filtering | Monte Carlo Methods | Linear Algebra | Machine Learning | Deep Learning | Hidden Markov Models | Reinforcement Learning | Stochastic Processes | Quantum Computation | Stochastic Differetial Equations | Mathematical Modeling | Probability and Statistics | Nonlinear Dynamics
Technical competency : Linux | Python | Latex | html
Frameworks and Libraries : Pytorch | Tensorflow | Jax | scipy | sklearn | Hugo | git version control
Lincences and Certifications:
- Deeplearning -Neuromatch Academy View
- Imperial College London-Online course on Data Assimilation view
- NVIDIA DLI Certificate - Applications of AI for Anomaly Detection View
- NVIDIA DLI Certificate - Fundamentals of Deep Learning View
- NVIDIA DLI CERTIFICATE -Accelerating Data Engineering Pipelines View
- IBM CERTIFIED ASSOCIATE DEVELOPER - Quantum Computation View
- IBM Quantum Challenge 2021 Achievement - Advanced View
A CV is the evidence that ..
Life must be understood backwards but lived forwards.
— Søren Kierkegaard (a 19th-century Danish philosopher)