My Sunday Adventures
The Sunday at INFORMS began with myself waking up too early, given the time I went to bed the night before (no, I wasn’t partying). To my surprise, the Starbucks at the Westin was empty and I had no trouble obtaining my morning dose of caffeine. It was right then that someone approached me and said “Hi! I read your blog!”. It took me a few seconds to fathom what had just happened and I said “Really? That’s great!”. We chatted for a little while and I learned his name was Hugh Medal, a PhD student from the University of Arkansas, whom I had never met before. At least now I know he works on network problems with disruptions. What a great way to start an INFORMS Sunday! Thanks, Hugh!
I had a research meeting at 10am, so I tried to squeeze in a couple of talks before that. Unfortunately, I completely overestimated my knowledge of Stochastic Programming when I assumed I’d be able to follow Suvrajeet Sen’s talk on “What is Common and What is Not Between Differential DP, Nested Benders and Stochastic Decomposition”. Not to worry, though. I learned I need to learn more. That’s still learning, right?
The second talk I had picked for the morning slot was “New Media Planning Models for New Media” by John Turner from UC Irvine. It was a fascinating application of OR to a real-world problem: how to serve ads to people playing video games while respecting a number of constraints on campaign goals set by the product manufacturers. Next time you see soda X on a vending machine of a shooting game, or politician Y on a billboard of a driving game, rest assured that those ads were strategically placed there to target a specific demographic, during a specific time of the year. John and his co-authors achieved dramatic reductions in penalty costs (which are charged when goals aren’t met), while also improving how well the campaign reaches its weekly target demographics. Even though their method requires solving large-scale quadratic transportation problems, they employ a smart aggregation procedure to significantly reduce the problem size and make it manageable. They’re now working on generalizing their approach to other kinds of digital advertisement.
Between then and 5pm I had two very interesting (and long) research meetings with some of my collaborators. The goal of the first meeting was to come up with a list of problems we think are worth studying; and the second meeting aimed at deciding which of those problems we want to tackle first. If only I could find more than 24 hours a day… All in all, I’m happy with the outcomes of those meetings and I’m excited to start playing around with Multi-Valued Decision Diagrams (MDD). In case you don’t know what MDDs are, I suggest you attend the MC08 session I’m chairing on Monday. Fascinating stuff! On a sad note, that means I’ll have to miss the panel on “Bringing OR into the 21st Century with Social Networking and Web 2.0 Tools” (MC03). Someone please let me know what happens in that session! (e.g. send me a tweet :-)
At 6:15pm I attended the INFORMS Optimization Society business meeting and spoke for a few minutes about the upcoming IOS meeting that will take place at the University of Miami in February next year. Yes, you read it correctly: Miami+February = nice! (At least for those who live in the northern heimsphere.) This is my first time helping organize a conference, and I must say it’s shaping up to be a very nice one. Make sure to check out our web site http://bus.miami.edu/ios. We will be accepting abstract and poster submissions until January 6, 2012. I left some brochures about the conference on the appropriate table in the Exhibits Hall. Feel free to take one and show it to all of your friends.
That’s about it for my exciting Sunday adventures. Looking forward to finding out what Monday has in store for me. One thing I know already: I’ll be waking up early once again because my first research meeting starts at 7:30am. Hopefully, another one of my blog readers will be walking by to cheer me up.
P.S. If you’re vegetarian, go to the Mimosa Grill and ask the chef for a “vegetarian creation” (not on the menu; they’ll make it on the spot). You won’t regret it.