|
I am a Postdoctoral Research Associate in the National Center for Computational Science Division at Oak Ridge National Lab. I am supported through the DOE's Leadership Computing Facility Distinguished Postdoctoral Program. I obtained my Ph.D. in Computer Science from Kansas State University advised by Dr. William Hsu. My research interest centers around deep reinforcement learning (RL), generative AI, high-performance computing (HPC), and scientific computing. During my graduate study, I had the opportunity to intern at DEKA Research and Development as a Machine Learning Intern and at C2FO as a Data Scientist Intern. Previously, I obtained an MS in Computer Science from Kansas State University in 2019 and I completed my Bachelor of Science (B.Sc.) in Computer Science & Engineering (CSE) from Bangladesh University of Engineering & Technology (BUET) in 2015. After graduation, I worked at REVE Systems Ltd. as a Jr. Software Engineer. Email / Full CV / Google Scholar / GitHub / Twitter / LinkedIn |
|
|
|
|
My research interest broadly lies in the intersection of Generative AI and Reinforcement Learning (RL). My research focuses on developing efficient, scalable, and generalizable machine learning systems that can be reliably deployed in complex tasks. I design learning frameworks that can effectively learn rich patterns and decision-making from intricate, high-dimensional data arising in physics-based and imaging-based domains. I believe effective representation learning, generative modeling, and scalable optimization will take us forward. I am interested to investigate the delicate balance between exploration and exploitation in preference alignment, active learning, and RL optimization. I am also excited to explore the role of memory (transformer like architecture) and transfer learning (from large models) in promoting generalization. I aim to push the boundaries of what AI agents can achieve by enabling robustness and generalization, unlocking their potential to operate in complex, diverse, and dynamic real-world environments. My previous works include RL generalization, efficient neural architecture design (transformer/CNN models), image/video classification, image synthesis using generative adversarial network (GAN) & diffusion models, meta-learning, domain adaptation, and distributed computing. In terms of application, my research contributed to intelligent agents, multi-agent system, agriculture (plant disease detection, pesticide spraying robot), sports safety (risky tackle detection), crisis-response (summarization of disaster news), and diverse scientific domains such as climate science, computational fluid dynamics, materials imaging. Please visit my google scholar profile for detailed list of publications. I am currenly looking for tenure-track faculty or research scientist roles. Please reach out if you think I will be a good fit for your open position. Thanks for visiting my site. |
|
|