Aishik Deb

PhD Candidate in Computer Science

Stony Brook University, NY, USA

Reinforcement Learning | Scientific Visualization | Visual Analytics

Get In Touch Download CV
Aishik Deb

About Me

I am a fourth-year Ph.D. candidate in Computer Science at Stony Brook University, advised by IEEE Fellow Prof. Klaus Mueller in the Visual Analytics and Imaging Lab. My research focuses on Reinforcement Learning and Scientific Visualization, with particular emphasis on developing novel exploration strategies and interactive tools for scientific discovery.

Research Lab: Visual Analytics and Imaging Lab
Advisor: Prof. Klaus Mueller



Research Interests

Reinforcement Learning

Optimizing Reinforcement Learning through Empty Space Search, developing novel exploration strategies that achieve superior performance in complex continuous control tasks.

Scientific Visualization

Building multi-resolution Transformer pipelines for real-time analysis of SAXS images and creating interactive visualization systems for scientific data exploration and discovery.

Visual Analytics

Developing interactive tools and systems that enable scientists to explore, analyze, and understand complex multivariate datasets more efficiently.

Publications

Gradient Efficient On-Policy Reinforcement Learning via Exploration of Sparse Parameter Space

X. Zhang, A. Deb, and K. Mueller

International Conference on Machine Learning (ICML) 2026

Under Review

Multivariate Volume Data: Advances in Visualization and Analysis Techniques

P. Avery, S. Jhaveri, C. Tsolakis, S. Jourdain, A. Deb, et al.

Microscopy and Microanalysis 2025, vol. 31, no. Supplement1, ozaf0481150

Published

Experience

Graduate Research Assistant

2023 - Present

Stony Brook University • Dr. Klaus Mueller

  • Optimizing Reinforcement Learning through Empty Space Search
  • Developing multi-resolution transformer pipelines for real-time SAXS morphology discovery
  • Built interactive multivariate data visualization systems

Graduate Research Assistant

2022 - 2023

Stony Brook University • Dr. Shubham Jain

  • Developed Time Series Prediction model for bench press failure prediction
  • Computed novel motion-derived metrics for muscle failure indication

Systems Engineer

2019 - 2022

Tata Consultancy Services • Kolkata, India

  • Led enterprise web migration from WEM to AEM serving 100k+ users, improving page performance by about 30% and deployment frequency by 2x via CI/CD automation
  • Built full-stack features across Angular/React, Node.js, PostgreSQL; containerized services with Docker and automated releases with Azure DevOps
  • Mentored 4 junior developers and received Special Initiative Award for project leadership

Data Science Intern

2018

Indian Statistical Institute • Kolkata, India

  • Implemented Map-Reduce Copy Number Variation (CNV) detection to parallelize genome segmentation, cutting processing time by about 25% across multiple DNA samples.
  • Developed a Flask + Apache Thrift interface over HBase to query terabytes of genomic data, reducing query time by about 40%.

Web Development Intern

2018

Oil and Natural Gas Corporation • Kolkata, India

  • Designed and implemented a Vehicle Requisition System for 500+ ONGC officers, reducing vehicle scheduling and approval time by 60% and streamlining management of official vehicle requests and allocations.
  • Built full-stack features using PHP, JavaScript, and MySQL, including secure login, request tracking, and automated notifications.

Education

Ph.D. in Computer Science

2022 - Present
Stony Brook University, New York, USA

GPA: 3.96/4.0

B.Tech. in Information Technology

2015 - 2019
Maulana Abul Kalam Azad University of Technology (MAKAUT), Kolkata, India

GPA: 8.9/10.0

Selected Projects

FairGPT: Examining the Intersectional Bias in GPT-3.5

2023 • Dr. Andrew Schwartz

Processed 22 GB of Wikipedia data, applied BERT NER with PySpark, trained BiLSTM for gender classification, and generated 80k name-conditioned stories using OpenAI API to identify intersectional biases in GPT-3.5.

Python PySpark BERT BiLSTM GCP

Life Expectancy Analysis Dashboard

2023 • Dr. Klaus Mueller

Built interactive D3.js Flask dashboard covering 183 countries across 16 years with synchronized charts, brushing/linking, and cross-filtered exploration capabilities.

D3.js Flask Python JavaScript

Awards & Honors

Contact

Email

aideb [at] cs [dot] stonybrook [dot] edu

Department

Department of Computer Science
Stony Brook University
New York, 11790

GitHub LinkedIn Google Scholar