Building Agile Machine Learning Models
Richard Zemel is a Professor and Machine Learning Research Chair in the Department of Computer Science at the University of Toronto. He will be joining the faculty in the Computer Science Department at Columbia University this summer. He is a Co-Founder and the Research Director of the Vector Institute for Artificial Intelligence. His awards include an NVIDIA Pioneers of AI Award, an ONR Young Investigator Award, a CIFAR AI Chair, and two NSERC Discovery Accelerators. Richard is on the Advisory Board of the Neural Information Processing Society. Professor Zemel current research interests include learning useful representations of data without any supervision, algorithmic fairness, learning with little data, and graph-based machine learning. His research is supported by grants from NSERC, CIFAR, Microsoft, Google, Samsung, DARPA, and iARPA.
The last few years have seen significant advances and real-world applications in machine learning and artificial intelligence, which have also led to a myriad of challenges. In this talk I will focus on three fundamental issues that confront our attempts to build flexible and effective ML systems. How can new tasks be learned readily, with relatively few examples? How can these systems maintain strong performance as the environment they are operating in differs from the training environment? I will discuss current approaches to these problems, and some surprising links to a third problem: how can these systems avoid discriminating against individuals or groups? I will highlight their successes and also current limitations and open problems.
