Ryan Adams

Ryan  Adams

Ryan Adams

Assistant Professor of Computer Science, Harvard School of Engineering and Applied Sciences


Symposium: Weathering the Data Storm: The Promise and Challenges of Data Science

Friday, January 24 

Title: Taking Humans Out of the Machine Learning Loop
Abstract: Recent years have seen significant methodological advances in machine learning tools for data analysis and reasoning under uncertainty. These tools have had an impact in a diverse range of domains, from visual object recognition to information retrieval to medical diagnosis. In many ways, machine learning is moving away from the long-term objective of developing intelligent algorithms to perform tasks automatically, as these modern tools require a great deal of human sophistication and insight. This need for expertise makes the results from the literature less reproducible and ultimately limits the impact that methodological innovations can have on society. In this talk, I will discuss my group’s work in developing methods to automate other machine learning and data analysis tools, and our work to take humans out of the loop of these problems. Remarkably, these automated procedures often outperform their human counterparts, even in very specialized domains. I will also describe how these techniques are also enabling new solutions to problems outside of computer science.

Bio: Ryan
Adams is an assistant professor of computer science in the Harvard School of Engineering and Applied Sciences. Adams leads the Harvard Intelligent Probabilistic Systems (HIPS) group, which focuses on probabilistic approaches to building machine learning algorithms, working at the interface of computer science, statistics and computational neuroscience. In broad terms, he is interested in understanding the computation that lies beneath intelligence and developing artificial systems that can discover complex structure in data. Adams completed his PhD in physics under David MacKay at the University of Cambridge, where he was a Gates Cambridge Scholar. His doctoral work won the honorable mention for the Savage Award for best dissertation in Bayesian theory and methods from the International Society for Bayesian Analysis. Before coming to Harvard, he spent two years as a Junior Research Fellow at the University of Toronto as a part of the Canadian Institute for Advanced Research.