Curriculum Vitae
Research Profile
Researcher working at the intersection of Scientific Machine Learning, Bayesian Inverse Problems, Uncertainty Quantification, Computational Fluid Dynamics, and Multiphysics Modeling.
My research combines high-fidelity physics-based simulations with machine learning, deep generative models, explainable artificial intelligence, and optimization to address forward and inverse problems in computational engineering.
My doctoral research focused on charged particle dynamics and electrostatic collection using multiphysics simulations, interpretable machine learning, and surrogate-assisted multi-objective optimization. My current research focuses on deep learning-based Bayesian inverse problems and uncertainty quantification.
Research Interests
- Scientific Machine Learning
- Bayesian Inverse Problems
- Uncertainty Quantification
- Probabilistic Machine Learning
- Deep Generative Models
- Computational Fluid Dynamics
- Multiphysics Modeling
- Explainable Artificial Intelligence
- Surrogate Modeling
- Multi-Objective Optimization
- Scientific Computing
Research Experience
Project Associate (Ad Hoc)
Indian Institute of Technology Palakkad
Department of Mechanical Engineering
February 2026 – Present
Mentor: Prof. Arijit Hazra
Working on deep learning-based methodologies for inverse problems and uncertainty quantification in computational engineering.
Current research activities include:
- Developing deep learning-based inverse modeling frameworks for complex physical systems.
- Investigating variational and generative approaches for Bayesian inverse problems.
- Developing computationally efficient approaches for posterior approximation.
- Comparing deep learning-based posterior estimates with sampling-based Bayesian methods.
- Investigating uncertainty quantification and posterior covariance estimation.
- Studying information geometry and Fisher Information Matrix-based characterization of inverse problems.
Doctoral Researcher
Indian Institute of Technology Delhi
Department of Mechanical Engineering
January 2020 – February 2026
Ph.D. Supervisors: Prof. Bahni Ray and Dr. Mayank Kumar
Conducted doctoral research integrating multiphysics simulations, computational fluid dynamics, electrostatics, Lagrangian particle tracking, machine learning, explainable AI, and multi-objective optimization for charged aerosol particle collection.
Major research contributions included:
- Development of high-fidelity three-dimensional multiphysics models for charged particle transport.
- Experimental validation of numerical particle collection models.
- Investigation of electrostatic, hydrodynamic, and particle rebound mechanisms.
- Development of machine learning surrogate models for computationally efficient prediction.
- Application of SHAP, PDP, and other explainability methods for identifying dominant physical parameters.
- Development of clustering-guided design-space reduction methodologies.
- Integration of surrogate models with NSGA-II for multi-objective engineering optimization.
Education
Doctor of Philosophy (Ph.D.) — Mechanical Engineering
Indian Institute of Technology Delhi
January 2020 – February 2026
Specialization: Fluid Mechanics and Thermal Sciences
GPA: 8.79/10
Thesis:
Modeling Charged Particle Dynamics and Optimizing its Collection Efficiency by Spherical Collector Using Machine Learning
Supervisors: Prof. Bahni Ray and Dr. Mayank Kumar
Master of Technology (M.Tech.) — Maintenance Engineering
Maulana Azad National Institute of Technology Bhopal
July 2017 – May 2019
GPA: 8.16/10
Bachelor of Technology (B.Tech.) — Mechanical Engineering
Dr. A.P.J. Abdul Kalam Technical University, Uttar Pradesh
August 2011 – June 2015
GPA: 7.08/10
Publications
Journal Articles
A. Srivastava, D. Biswal, B. Ray, M. Kumar, and D. Dasgupta,
“Influence of nonuniform electric field due to a spherical collector on the particles’ dynamics in quiescent air,”
Journal of Electrostatics, 129, 103916, 2024.
A. Srivastava, S. Kandpal, B. Ray, M. Kumar, and D. Dasgupta,
“Analyzing the Impact of Charge Particle Rebound on Spherical Collector Efficiency via Explainable Machine Learning,”
Powder Technology, 469, Part 2, 121803, 2025.
Under Review / Submitted
A. Srivastava, S. Kandpal, B. Ray, M. Kumar, and D. Dasgupta,
“Machine learning-based regime identification and size-resolved optimization of charged-particle capture and rebound on spherical collectors.”
Powder Technology (submitted).
A. Srivastava, A. Hazra, and R. Dubaku,
“Conditioning-dependent variational inference for uncertainty quantification in physics-based inverse problems.”
Probabilistic Engineering Mechanics (under review). arXiv:2609.25145
Conference Contributions
A. Srivastava, D. Biswal, B. Ray, D. Dasgupta, R. Thaokar, and Y. S. Mayya,
“Electrospray Based Air Pollution Mitigation: Droplet and Particle Interaction.”
COMPOLL: Joint Conference on Pollution Awareness in Research and Teaching, IIT Delhi, November 2022.
A. Srivastava, B. Ray, M. Kumar, D. Dasgupta, R. Thaokar, and Y. S. Mayya,
“Dynamic Characteristics of Submicron Particle Deposited on the Charged Spherical Collector.”
Conference on Fluid Mechanics and Fluid Power (FMFP), IIT Roorkee, December 2022.
A. Srivastava, B. Ray, and M. Kumar,
“Numerical Investigation on Deposition Process of Polydisperse Dust Particles on Charged Spherical Pendant Collector.”
International Workshop on Sustainable Energy, Power and Propulsion (IWSEPP), IIT Kanpur, January 2024.
Selected Research Projects
Bayesian Deep Learning for Inverse Problems and Uncertainty Quantification
Development of deep learning and variational methodologies for Bayesian inverse problems with emphasis on posterior approximation, uncertainty quantification, and information geometry.
Clustering-Guided Surrogate-Assisted Multi-Objective Optimization
Development of an optimization framework integrating machine learning surrogate models, K-means clustering, NSGA-II, TOPSIS, and explainable AI.
Explainable AI for Charged Particle Collection
Development of interpretable machine learning frameworks using LightGBM, SHAP, PDP, and ALE to understand charged particle collection and rebound mechanisms.
Multiphysics Modeling of Charged Particle Dynamics
Development of three-dimensional CFD–electrostatics–Lagrangian particle tracking models for investigating charged aerosol particle transport and deposition.
Teaching Experience
Teaching Assistant — Intermediate Heat Transfer
Indian Institute of Technology Delhi
Teaching Assistant — Advanced Thermodynamics
Indian Institute of Technology Delhi
Relevant Graduate Coursework
- Computational Heat Transfer
- Advanced Fluid Mechanics
- Advanced Heat Transfer
- Microfluidics
Awards & Honors
Best Technical Interaction Award — IWSEPP 2024
Received the Best Technical Interaction Award at IWSEPP-2024 for technical engagement and scientific discussions.
MHRD Fellowship / Institute Assistantship
Received doctoral fellowship/assistantship support during Ph.D. studies at the Indian Institute of Technology Delhi.
Graduate Aptitude Test in Engineering (GATE)
Qualified GATE 2016 — Mechanical Engineering with 97.93 percentile among approximately 200,000 candidates.
Ph.D. Admission Offers
Received Ph.D. admission offers from IIT Delhi, IIT Kanpur, and IIT Kharagpur in 2019.
Academic Leadership
Research Scholar Representative
Student Affairs Council, IIT Delhi
June 2023 – May 2024
Served as a Research Scholar Representative in the Student Affairs Council (SAC), IIT Delhi.
Technical Skills
Programming & Scientific Computing:
Python, NumPy, SciPy, Pandas
Scientific Machine Learning:
Scikit-learn, LightGBM, Regression, Classification, Clustering, Surrogate Modeling
Deep Learning & Probabilistic Learning:
TensorFlow, Keras, Variational Autoencoders, Variational Inference
Explainable Artificial Intelligence:
SHAP, Partial Dependence Plots (PDP), Accumulated Local Effects (ALE)
Bayesian & Uncertainty Quantification:
Bayesian Inference, Bayesian Inverse Problems, MCMC, Variational Inference, Fisher Information Matrix, Posterior Analysis
Computational Engineering:
Computational Fluid Dynamics, Multiphysics Modeling, Finite Element Modeling, Lagrangian Particle Tracking
Simulation Software:
COMSOL Multiphysics, FEniCS
Optimization:
Genetic Algorithms, NSGA-II, Multi-Objective Optimization
Visualization:
Matplotlib, Plotly
Development & Research Tools:
Git, GitHub, VS Code, LaTeX, Jupyter, Linux
Professional Profiles
Google Scholar · ORCID · GitHub · ResearchGate · LinkedIn