《Unbiased and Optimal Learning to Rank
Abstract:
Consider the scenario where an algorithm is given a context, and then it must select a slate of relevant results to display. As four special cases, the context may be (i) a search query, (ii) a slot for an advertisement, (iii) a social media user, or (iv) an opportunity to show recommendations. We want to compare many alternative ranking functions that select results in different ways. However, A/B testing with traffic from real users is expensive. This talk will present a novel method that reuses traffic that was exposed to a past ranking function to estimate the utility of a hypothetical new ranking function. The method is a purely offline computation, and relies on assumptions that are quite reasonable in practice. We show further how to design a ranking function that is the best possible, given the same assumptions. Learning optimal rankings for search results given queries is a special case. Experimental findings on data logged by a real-world e-commerce web site are positive.
Bio:
Charles Elkan is an adjunct professor of computer science at the University of California, San Diego (UCSD). His full-time position is as managing director and global head of machine learning at Goldman Sachs. From 2014 to 2018 he was the first Amazon Fellow, leading a team of over 30 scientists and engineers in Seattle, Palo Alto, and New York doing research and development in applied machine learning for both e-commerce and cloud computing. Before joining Amazon, Dr. Elkan was a tenured full professor of computer science at UCSD. His Ph.D. is from Cornell and his undergraduate degree from Cambridge, and he has held visiting positions at Harvard and MIT. His students have gone on to faculty positions at universities including Columbia, Carnegie Mellon, the University of Washington, and Stanford, and to leading roles in industry.
《Making computer vision systems that work: Boujou, Kinect, HoloLens
Abstract:
I have been lucky enough to have been involved in the development of real-world computer vision systems for over twenty years. In 1999, prize-winning research from Oxford University was spun out to become the Emmy-award-winning camera tracker “boujou”, which has been used to insert computer graphics into live-action footage in pretty much every movie made since its release, from the “Harry Potter” series to “Bridget Jones’s Diary”. In 2007, I was part of the team that delivered human body tracking in Kinect for Xbox 360, and in 2015 I moved from Microsoft Research to work on Microsoft’s HoloLens, an AR headset brimming with cutting-edge computer vision technology, where I worked on the fully-articulated hand tracking for HoloLens 2. In all of these projects, the academic state of the art has had to be leapfrogged in accuracy and efficiency, sometimes by orders of magnitude. Sometimes that’s just raw engineering, sometimes it means completely new ways of looking at the research. If I had to nominate one key to success, it’s a focus on, well, everything: from low-level coding to algorithms to user interface design, and on always being willing to change one’s mind.
Bio:
Andrew Fitzgibbon leads the “All Data AI” (ADA) research group at Microsoft in Cambridge, UK.
He is best known for his work on 3D vision, having been a core contributor to the Emmy-award-winning 3D camera tracker “boujou“, to body tracking for Kinect for Xbox 360, and for the articulated hand-tracking interface to Microsoft’s HoloLens.
His research interests are broad, spanning computer vision, machine learning, programming languages, computer graphics and occasionally a little neuroscience.
He has published numerous highly-cited papers, and received many awards for his work, including ten “best paper” prizes at various venues, the Silver medal of the Royal Academy of Engineering, and the BCS Roger Needham award. He is a fellow of the Royal Academy of Engineering, the British Computer Society, and the International Association for Pattern Recognition, and is a Distinguished Fellow of the British Machine Vision Association.
Before joining Microsoft in 2005, he was a Royal Society University Research Fellow at Oxford University, having previously studied at Edinburgh University, Heriot-Watt University, and University College, Cork.
《Fairness and Bias in Algorithmic Decision-Making
Abstract:
Recent discussion in the public sphere about classification by algorithms has involved tension between competing notions of what it means for such a classification to be fair to different groups. We consider several of the key fairness conditions that lie at the heart of these debates, and discuss recent research establishing inherent trade-offs between these conditions. We also explore how the complexity of a classification rule interacts with its fairness properties, showing how natural ways of approximating a classifier via a simpler rule can act in conflict with fairness goals.
In this talk, I will start by introducing a very general form of variable importance, called model class reliance. Model class reliance measures how important a variable is to any sufficiently accurate predictive model within a class. I will use this and other data-centered tools to provide our own investigation of whether COMPAS depends on race, and what else it depends on. Through this analysis, we find another problem with using complicated proprietary models, which is that they seem to be often miscomputed. An easy fix to all of this is to use interpretable (transparent) models instead of complicated or proprietary models in criminal justice.
The talk will be based on joint work with Jens Ludwig, Sendhil Mullainathan, Manish Raghavan, and Cass Sunstein.
Bio:
Jon Kleinberg is the Tisch University Professor in the Departments of Computer Science and Information Science at Cornell University. His research focuses on the interaction of algorithms and networks, and the roles they play in large-scale social and information systems. He is a member of the National Academy of Sciences and the National Academy of Engineering, and the recipient of MacArthur, Packard, Simons, Sloan, and Vannevar Bush research fellowships, as well awards including the Harvey Prize, the Nevanlinna Prize, and the ACM Prize in Computing.
《Unified data + unified computation = Multi-paradigm data science
Abstract:
While greater automation has made machine learning and data science tools accessible to non-experts, that same automation is equally important to the expert user. By breaking down the barriers between different kinds of computation a truly multi-paradigm approach to data science becomes possible.
This talk will demonstrate Wolfram Research's progress towards a fully unified computation platform including live coded machine-learning, computer vision and production deployment.
Making this all possible is an underlying symbolic representation that unifies data, models, code and interfaces. The talk will explain how this simplifies high-level concepts and enables their automation. Examples will include surgery and transfer learning on a neural network and automated anomaly detection.
Bio:
As Director of Technical Services, Communication and Strategy at Wolfram Research Europe, Jon McLoone is central to driving the company's technical business strategy and leading the consulting solutions team. Described as “The Computation Company”, the Wolfram group are world leaders in integrated technology for computation, data science and AI including machine learning. With over 30 years of experience working with Wolfram Technologies, Jon has helped in directing software development, system design, technical marketing, corporate policy, business strategies and much more. Jon gives regular keynote appearances and media interviews on topics such as the Future of AI, Enterprise Computation Strategies and Education Reform, across multiple fields including healthcare, fintech and data science. He holds a degree in mathematics from the University of Durham. Jon is also Co-founder and Director of Development for computerbasedmath.org, an organisation dedicated to a fundamental reform of maths education and the introduction of computational thinking. The movement is now a worldwide force in re-engineering the STEM curriculum with early projects in Estonia, Sweden and Africa.
