Projects

Reinforcement Learning for Fuzzing

Reinforcement Learning, Fuzzing

Developed RL-based models to generate test cases, significantly improving vulnerability detection and code coverage over traditional fuzzing techniques.

AI-Based Anomaly Detection in Cyber-Physical Water Testbeds

Machine Learning, Anomaly Detection

Working on machine learning models for real-time anomaly detection in water systems, enhancing security and resilience of cyber-physical infrastructure.

Image Synthesis and Reconstruction

Diffusion Models, Computer Vision

Used Denoising Diffusion Probabilistic Model on Amsterdam Library of Textures (ALOT) dataset to generate diverse, high-fidelity artificial textures resembling originals.

Deep Learning-based Steganography

Deep Learning, Steganography

Implemented a framework to securely embed one image within another, ensuring high imperceptibility and robust recoverability of hidden data.

Latent Diffusion Models for High-Resolution Image Synthesis

Latent Diffusion Models, Text-to-Image

Implemented LDMs for efficient text-to-image generation by applying diffusion in latent space using pretrained autoencoders. Integrated cross-attention with U-Net for conditional synthesis.

Improving Lane Level Dynamics for EV Traversal: A Reinforcement Learning Approach

Tyagi, A., Lowalekar, M. and Paruchuri, P., 2024. In VEHITS (pp. 134-143). DOI: 10.5220/0012637200003702. Nominee for Best Student Paper Award.

A Reinforcement Learning Approach to Multi-Parametric Input Mutation for Fuzzing

Uwibambe, M.L., Tyagi, A. and Li, Q., 2025. In 2025 IEEE International Conference on Cyber Security and Resilience (CSR).