Areas of Interest

• Artificial Intelligence
• Control, Intelligent Systems, and Robotics
• Signal Processing
• Probabilistic Inference and Learning

Poster

Multi-Agent Active Search and Rescue

Abstract

In this poster, I propose a parallel active learning algorithm called SPATS (Sparse Parallel Asynchronous Thompson Sampling) that using multiple aerial and/or ground robots (agents) efficiently locates survivors of disasters. Unlike existing algorithms, SPATS is a practical algorithm that takes into account sparsity, lack of reliable communication to a central unit and sensing action constraints.

Bio

I am a postdoctoral fellow at the Robotics Institute at Carnegie Mellon University working with Prof. Jeff Schneider. My research interests include probabilistic inference and estimation, active learning, sparse signal recovery, Bayesian Optimization and machine learning. Prior to CMU, I completed my PhD at Cornell University with Prof. Christoph Studer developing novel inference, estimation and initialization algorithms for applications in machine learning, imaging and wireless communications. At Cornell, I was the recipient of the 2013-2014 Cornell fellowship and 2014 Jacobs Scholarship. Prior to Cornell, I obtained my Bachelor's degree in Electrical Engineering from Sharif University of Technology in Iran. Throughout the years, I have been involved in a variety of outreach activities supporting women in STEM led by organizations such as Society of Women Engineers, Girls Who Code and CATALYST Academy. Last year, I designed and led my own workshop to promote Electrical Engineering at Cornell’s largest workshop for middle school girls called Expanding Your Horizon.

Personal home page