One of the promises of artificial intelligence (AI) and machine learning is to assist scientists in dealing with the unprecedented scale and complexity of modern scientific datasets, particularly in the biomedical sciences. This talk will explore steps towards achieving this goal for AI, focusing on genomics and its applications in personalized medicine. In the first part of the talk, I will argue that the cost of scientific experiments can be dramatically reduced using statistical techniques. I will introduce a new genome sequencing technology that reduces the cost of genome haplotyping by up to tenfold by augmenting standard sequencing methods with a statistical model of the genome. This technology currently powers the phased sequencing product of Illumina Inc. In the second part of the talk, I will explore how AI can help synthesize scientific knowledge from the academic literature. I will present GWASkb, the largest automated curation effort aimed at studies linking genetics with human traits. GWASkb finds genetic associations that human curators miss and makes them available for disease risk prediction and basic research. The last part of the talk will focus on transforming scientific knowledge into medical decisions. I will describe safe and reliable machine learning systems that estimate disease risk using probabilistic forecasts that are accurate and capture calibrated uncertainty even when patient data is not i.i.d., but is rather chosen by a malicious adversary. The techniques described in this talk will help scientists make new discoveries and turn them into personalized medical technologies.

Volodymyr Kuleshov is a post-doctoral researcher in the group of Stefano Ermon in the Department of Computer Science at Stanford University. He obtained his Ph.D. in Computer Science from Stanford University, advised by Serafim Batzoglou and Michael Snyder. His research interests lie at the intersection of machine learning and computational genomics, and focus on two high-level goals: building intelligent tools that accelerate scientific discovery in biomedicine and developing core machine learning techniques that make such tools possible. Volodymyr has been awarded an NSERC Post-Graduate Fellowship and a Stanford Graduate Fellowship. His research has been featured in Nature Biotechnology, Scientific American, GenomeWeb, and Science Daily. His online lecture notes on probabilistic graphical models have been viewed more than 125,000 times by over 28,000 readers across the world and are used at Stanford. Part of his research on statistical genome phasing has been licensed commercially and now powers the phased sequencing product of Illumina Inc, following its acquisition of the startup Moleculo.