Paulina Varshavskaya

What I'm up to now --- Pacific Northwest Science

These pages are out of date as of July 2010.

I am currently visiting Sethu Vijayakumar's Statistical Machine Learning and Motor Control group at the School of Informatics, University of Edinburgh.

During the 2008-2009 academic year, I was a member of the MIT Center for Collective Intelligence, where I pursued research on multi-agent machine learning and distributed decision-making. In Spring and Summer of 2008, I was a Postdoc in Daniela Rus's Distrbuted Robotics Lab, working on distributed reinforcement learning for groups of robots such as self-reconfiguring modular robots or robotic swarms. Visit my research page for more details.

In the Spring of 2009, I am teaching COMP150-07 Intelligent Robotics at Tufts University.

In the Fall of 2008, I am teaching COMP131 Artificial Intelligence at Tufts University.

I am on the program committee of the 9th International Symposium on Distributed Autonomous Robotic Systems (DARS 2008). I co-organized the RSS 2007 Workshop on Algorithmic Equivalences Between Biological and Robotic Swarms with James McLurkin.

My article on self-reconfiguring modular robots appears in "Popularnaya mehanika" (Popular Mechanics Russia) in May 2007. I weigh in at The Boston Phoenix on the eternal question of whether robots will take over the world.

I was born in St-Petersburg, Russia and grew up in Central Europe (Budapest and Vienna). My undergraduate years were spent at University College London, studying computer science and cognitive science. My MS work at the MIT AI Lab, as it was then known, was on proto-word acquisition for the humanoid robot Kismet in Rod Brooks's lab. My PhD work with Daniela Rus and Leslie Pack Kaelbling was on reinforcement learning in distributed robotics. I graduated in July 2007. You can download my thesis titled Distributed Reinforcement Learning for Self-Reconfiguring Modular Robots. At some point it will be available through DSpace.

I am also interested in broader issues of AI, with a focus on decentralized and distributed intelligence, behavior-based robotics, biologically and neurally inspired mechanisms and computing, natural language processing and acquisition, machine learning, and philosophy of mind and AI.

I ran the Robotics Journal Club from the Summer of 2004 through Spring of 2007.

I have been seen around MITOC, climbing, books, pictures, go. I dislike writing about myself in the third person, or the first person plural, for that matter.

Russian authorities seem to have trouble spelling in a foreign alphabet, and therefore my name has been written at various times as Paulina Varchavskaia, Paulina Varshavskaya or Polina Varshavskaya. Some time ago Hungarian authorities contributed Polina Varsavszkaja. In a former francophile life I was also known as Pauline Varchavsky. So if you are looking for any of the above, it might be me.