I am software developer and machine learning researcher living in Toronto, Canada. I recently completed a postdoc at Oregon State University working with Tom Dietterich. I received by Ph.D. in Computer Science at the University of British Columbia with David Poole.
My research explores optimization algorithms appropriate for problems with spatial structure, particularly in computational sustainability domains. My thesis introduced a policy gradient optimization algorithm called Equilibrium Policy Gradients(EPG) for spatiotemporal domains. EPG utilizes a parametrized policy defined as the equilibrium distribution of a Markov chain built from local causal policies. The most recent paper on this work can be found here.
Other Interests: machine learning (big data, data analytics), reinforcement learning, causal modelling, Bayesian Networks, inference in probabilistic graphical models, game theory, mechanism design, object oriented software design, logic programming.
I am currently looking for exciting research and/or development opportunities in academia or industry. Feel free to contact me by email(contact@<this_domain>), on Google+(+Mark Crowley) or on Twitter(@compthink).
- Sept, 2014: I’m back in Toronto! Get in touch if you want to talk AI, Machine Learning, computational Sustainability or exciting software development projects.
- July 24, 2014: Colloqium Talk at the Department of Electrical and Computer Engineering at the University of Waterloo. Title: Big Data and Big Decisions – From Sustainable Management to Large Scale Spatiotemporal Decision Making.
- November 25, 2013: I’ll be giving a colloqium talk of the School of Electrical Engineering & Computer Science at Oregon State University.
- October, 2013 : Attending the 1st Multidisciplinary Conference on Reinforcement Learning and Decision Making (RLDM2013) in Princeton presenting my latest work on EPG planning in spatial domains with experiments on a wildfire fuel treatment simulator. See all the abstracts and papers for RLDM2013 here.
- September, 2013: Went kayaking with Humpback whales in the Great Bear Rainforest!
- August 5, 2013: Extended abstract was accepted for RLDM2013.
- July, 2013: Attending AAAI 2013 in Bellevue, Washington for the paper “PAC Optimal Planning for Invasive Species Management : Improved Exploration for Reinforcement Learning from Simulator-Defined MDPs” coauthored with Thomas Dietterich, Majid Alkaee Taleghan.
- July, 2013: Attending UAI 2013 in Bellevue, Washington.
- May, 2013: Attending the Canadian Operational Research Society (CORS2013) in Vancouver, BC
- February 25, 2013: Giving an invited talk on my research at the Computer Science Department at the University of Alberta in Edmonton.
- January, 2013: Gave a talk at the Informs Computing Society Conference in Sante Fe, New Mexico.
- December 2012: Attended NIPS2012 in Lake Tahoe, Nevada.