Atoosa Parsa

"Do not follow where the path may lead, go instead where there is no path and leave a trail."

-- Ralph Waldo Emerson

I am an AI researcher, working at the intersection of physics, biology, and computer science. Nothing excites me more than a challenging research question or an unconventional viewpoint on existing problems1. I like to employ Machine Learning techniques to investigate big questions about Embodied Intelligence and apply my findings to real-world problems. I have a broad set of interests, including:

I recently completed my Ph.D. in Computer Science (thesis) at the University of Vermont (UVM), where I was a member of the Morphology, Evolution, and Cognition Laboratory, and was advised by Professor Josh Bongard. Much of my Ph.D. research was focused on the unconventional computing paradigm, where I worked on creating Computational Granular Metamaterials (CGMMs). My work spanned devising conceptual frameworks for wave-based frequency-encoded computation, developing optimization pipelines for smart materials, and investigating their applications in the soft robotics field.

Recently, I’ve been exploring the underlying mechanisms of (poly)computation in CGMMs and studying the micro-macro relations that give rise to the desired computational capabilities using the modern Koopman theory (check out this page for details.). In addition, I’m currently engaged in a collaborative project with the Faboratory, where I work on the inverse design of a novel programmable robotic structure (see variable-stiffness particle configurations.).

Before joining UVM, I completed my M.Sc. degree in Machine Intelligence and Robotics at the University of Tehran and was advised by Professor Majid Nili Ahmadabadi. During this time, I worked on developing design methodologies to navigate the morphological space of compliant quadruped robots and adjust their structural parameters to improve performance (check out this link for details.). Before switching to the AI world, I was trained as a Computer Hardware Engineer at the University of Tehran. For my B.Sc. thesis, I worked on the hardware implementation of approximate computational blocks (see more details here.).

If you have similar interests and have a project that I can help with or collaborate on, do not hesitate to contact me.

1 Or finding an old (and forgotten?) article!