New Interactions Between Statistics and Optimization (22w5094)

The Banff International Research Station will host the "New Interactions Between Statistics and Optimization" workshop in Banff from May 22 - 27, 2022.

Machine learning is the process by which a computer autonomously extracts knowledge from data. After years of continual progress, machine learning is now able to carry out difficult tasks such as image classification, language translation, or genomic data analysis. It typically works as follows: first a human designs a computer program with many free parameters, and then an algorithm adjusts these parameters so that the program achieves a good performance on training data. The design and analysis of such algorithms is done by researchers in optimization. But the actual goal is to solve the task for new data that does not belong to the training set. Understanding when and why a good performance on training data also leads to a good performance on unseen data is the task of researchers in statistics.

While optimization and statistics are both relevant to advance machine learning, they are distinct research communities, with different tools and languages. In the past, important breakthroughs have occurred when these two communities have interacted, an example being the introduction of the Stochastic Gradient Descent algorithm, used to train most machine learning programs in today’s large scale applications. The objective of this workshop is to facilitate new interactions and collaborations between the statistics and optimization communities, with the intention of sparking new ideas at the interface of the two fields.

The Banff International Research Station for Mathematical Innovation and Discovery (BIRS) is a collaborative Canada-US-Mexico venture that provides an environment for creative interaction as well as the exchange of ideas, knowledge, and methods within the Mathematical Sciences, with related disciplines and with industry. The research station is located at The Banff Centre in Alberta and is supported by Canada's Natural Science and Engineering Research Council (NSERC), the U.S. National Science Foundation (NSF), Alberta's Advanced Education and Technology, and Mexico's Consejo Nacional de Ciencia y Tecnología (CONACYT).