131 reputation
2
bio website
location Mountain View, CA
age 27
visits member for 9 months
seen Sep 21 at 6:22

I work at Google, MapReduce team.

Previously, I've been at Mirantis (working on cluster computing and subsequently on cloud-related stuff) and at Yandex (Russia's leading search engine), doing information retrieval.

I'm interested in functional programming, application of mathematical structures to computer science and distributed systems. I'm a fan of classical music (especially Rachmaninoff, Scriabine and Bach) and occasionally play guitar and piano in my free time.


Sep
24
awarded  Autobiographer
Jan
25
comment Optimization / personalization within clusters
Yes. In the example I gave, evaluating $f$ would be changing the game parameters around a player and seeing how their engagement changes (it takes a long time, and if I pick really bad parameters, they may stop playing the game at all).
Jan
23
comment Optimization / personalization within clusters
Not necessarily runtime, but every measurement has a high cost (and still higher cost if I pick a bad value of P), so I'd like to minimize the number of measurements - naturally, reusing measurements between similar species of A would help greatly.
Jan
22
comment Optimization / personalization within clusters
The reason for that is explained in #2 : clustering, or some measure of similarity between points of A, would mean that I could optimize P within a whole cluster (at least I could get a very good approximation to the optimum), rather than at each individual point.
Jan
16
awarded  Student
Jan
16
asked Optimization / personalization within clusters