I'm not sure this'd even work... But let's see.
I nominate 22nd April Saturday, 3pm gather for coffee breaks in Sydney CBD.
Let's me briefly describe myself:
My background is in computing, finance and mathematics. I'm currently a bioinformatician, specalized in genomic analysis. I use statistics to develop models for our DNA genome.
I used to work as a quantitative analyst in finance, where I helped developing option pricing models. But I find statistics more interesting.
I'd argue myself an expert in chess programming. I write my own chess engines and sell them for profits on mobile platform (http://www.smallchess.com). I'm currently the leader in chess engine on chess.stackexchange.com by a wide margin (2x the second leader):
I'm applying statistics and machine learning in chess. I don't just use save statistics R-code somewhere on Github. I'd actually translate the code into C++, and sell it on the App Store.
Let's run through two example I'm still working on:
Q1: What's the piece value for a given chess position?
Chess players use their experience to say which chess pieces are "good", and which chess pieces are "bad". Imagine there is a mobile app that you can download from the App Store, where you give it a chess position and it tells you statistical value for each piece? What about the Monte Carlo tree search? Maybe we can use logistic regression?
Q2: How to derive similarity measure for two different chess positions?
This is actually an insane problem. The positions must have similar evaluation to be considered similar. Unfortunately, there is no known way to do it without running a chess engine. Running a chess engine is very slow, we would like to do it statically. Imagine there is a reliable similarity measure, we'd be able to apply machine learning on a given position, and compare it to a set of pre-trained models.
We can talk about anything interesting related to data analytics and chess!
EDIT (we have two):
I have got a taker. So we have two (including myself). More!
Bump up the page.