Event Date
SPEAKER: Minjie Fan; Google
TITLE: “Google Accelerated Science Overview”
ABSTRACT: In this talk, I will describe two research projects I have been working on in the research team called Google Accelerated Science, whose mission is to produce breakthroughs in natural sciences by applying Google technologies. The first project is neural-guided symbolic regression that incorporates semantic priors of expressions (such as leading powers) into expression finding tasks. The second project is phenotypic analysis of high-throughput cell screens that adapts pre-trained neural networks (trained on consumer images) to obtain embedding vectors from microscopic cell images for clustering drugs by their mechanism-of-actions and hopefully finding new drugs. In particular, we also developed a specific method for correcting nuisance variation such as batch effect for the embedding vectors.
DATE: Thursday, May 9th, 4:10pm
LOCATION: MSB 1147, Colloquium Room
REFRESHMENTS: 3:30pm MSB 4110 (4th floor lounge)
STA 290 Seminar List: https://statistics.ucdavis.edu/seminars