CV/Profile

Graduate Research Assistant

Department of Computer Science, George Mason University

Working on progressing the robotics frontier for planning under uncertainty.

• Robotics

Graduate Research Assistant

Department of Computer Science, Utah State University

Worked on research projects related to usable security and privacy.

• Human Computer Interaction

Lecturer

Department of Computer Science and Engineering, Islamic University of Technology

Conducted undergraduate courses, labs, and supervised undergraduate projects.

• Web Programming
• Mathematical Analysis
• Data Structures

Doctor of Philosophy

George Mason University

Computer Science

Master of Science

George Mason University

Computer Science

Bachelor of Science

Islamic University of Technology

Computer Science and Engineering

Fully Funded OIC Scholarship for Bachelor's Degree

Organization of Islamic Cooperation (OIC)

Awarded based on a competitive test.

Object Search in Partially-Known Environments via LLM-informed Model-based Planning and Prompt Selection

IEEE

We present a novel LLM-informed model-based planning framework, and a novel prompt selection method, for object search in partially-known environments. Our approach uses an LLM to estimate statistics about the likelihood of finding the target object when searching various locations throughout the scene that, combined with travel costs extracted from the environment map, are used to instantiate a model, thus using the LLM to inform planning and achieve effective search performance. Moreover, the abstraction upon which our approach relies is amenable to deployment-time model selection via the recent offline replay approach, an insight we leverage to enable fast prompt and LLM selection during deployment. Simulation experiments demonstrate that our LLM-informed model-based planning approach outperforms the baseline planning strategy that fully relies on LLM and optimistic strategy with as much as 11.8% and 39.2% improvements respectively, and our bandit-like selection approach enables quick selection of best prompts and LLMs resulting in 6.5% lower average cost and 33.8% lower average cumulative regret over baseline UCB bandit selection. Real-robot experiments in an apartment demonstrate similar improvements and so further validate our approach.

Anticipatory Planning for Performant Long-Lived Robot in Large-Scale Home-Like Environments

IEEE

This research addresses the challenge of robots performing sequential tasks in persistent, large-scale environments. Traditional planners often act myopically, optimizing only for the immediate task without considering future consequences. While anticipatory planning can improve efficiency by factoring in expected future costs, it typically struggles to scale in environments with numerous assets. To solve this, the authors introduce a scalable, model-based anticipatory task planning framework. By leveraging a graph neural network and a scene graph representation, the model learns key environmental properties to estimate future costs. Combined with a sampling-based procedure, this approach significantly reduces the overall cost of task sequences in various environments, especially when the robot is allowed to prepare in advance.

Active Information Gathering for Long-Horizon Navigation Under Uncertainty by Learning the Value of Information

IEEE

In this paper we address the task of long-horizon navigation in partially mapped environments for which active gathering of information about faraway unseen space is essential for good behavior. We present a novel planning strategy that, at training time, affords tractable computation of the value of information associated with revealing potentially informative regions of unseen space, data used to train a graph neural network to predict the goodness of temporally-extended exploratory actions. Our learning-augmented model-based planning approach predicts the expected value of information of revealing unseen space and is capable of using these predictions to actively seek information and so improve long-horizon navigation.

Improving Reliable Navigation Under Uncertainty via Predictions Informed by Non-Local Information

IEEE

This paper presents a novel approach to improve efficieny of reliable navigation with partial map by leveraging non-local information to inform predictions for good behavior.

Code to comment translation: A comparative study on model effectiveness & errors

ACL

This paper is a qualitative investigation into the various error modes of current state-of-the-art models for automated source code summarization. Instead of automatic reference metric for error, manual coding was used to analyze that qualitatively.

Understanding the sensibility of social media use and privacy with Bangladeshi Facebook group users

ACM

A qualitative study focused on understanding the sensibility of social media use and privacy of Bangladeshi Facebook group users.