ESP Biography
GABOR ANGELI, Stanford PhD student in AI (NLP)
|
Major: Computer Science College/Employer: Stanford Year of Graduation: G |
|
Brief Biographical Sketch:
A long long time ago, in a country far, far away, a young Hungarian lad made his way to Arizona, eventually wandering over to California to become a binary-tree-hugger at Berkeley. After which, he decided to combine his love of [syntax] trees and, apparently, school, and joined Stanford's Natural Language Processing group. OK, the syntax tree part is kind of a lie -- he actually spends most of his time working on semantics, most recently wrangling the internet to try to infer common-sense facts. In his free time, he likes board games, the outdoors, and building things. Oh, and cool hats. Past Classes(Clicking a class title will bring you to the course's section of the corresponding course catalog)M3093: Sampling, Genetic Algorithms, and Evolution in Splash! Fall 2013 (Nov. 02 - 03, 2013)
Sampling from a probability distribution is an essential component in tasks such as machine learning or simulation, and is often surprisingly difficult. For example: sampling animals according to how likely they are to survive (the analogy used in the class), sampling dynamic video game scenes according to how realistic they are, or helping a lost robot navigate by sampling over possible locations.
The theoretical portion of the class introduces the basics of sampling, and algorithms up through genetic algorithms and Markov Chain Monte Carlo methods. The second half of the class will be a programming project implementing the methods from the first half to make your own learning program!
|
|