Email: anna -at- harutyunyan -dot- net
Research keywords: reinforcement learning, reward shaping, off-policy learning, time series prediction
- [August 2016] Our paper Safe and Efficient Off-Policy Reinforcement Learning was accepted at NIPS!
- [January 2016] Our applied paper Predicting Seat-Off and Detecting Start-of-Assistance Events for Assisting Sit-to-Stand with an Exoskeleton was accepted for publication in IEEE Robotics and Automation Letters
- [September 2015] I will be spending this fall in London, interning at Google DeepMind
- [May 2015] Our papers Multi-Scale Reward Shaping via an Off-Policy Ensemble, Policy Transfer using Reward Shaping, and the demonstration Shaping Mario with Human Advice were presented at AAMAS 2015
I am a PhD student at the AI lab of VU Brussel, working with Prof. Ann Nowe. My research is primarily in reinforcement learning. I am interested in designing and analyzing efficient and scalable learning architectures, which requires many ingredients, such as reward shaping, ensembles and off-policy learning. I am also involved in an assistive exoskeleton project, where I am responsible for providing the high-level control, and where I do a lot of prediction.
Before this, I received my Masters degree at Oregon State University with the amazing Prof. Cora Borradaile, working on max flow algorithms in planar graphs.
My CV is can be found here, if you are into that kind of a thing.