Timezone: »

Simple and near-optimal algorithms for hidden stratification and multi-group learning
Christopher Tosh · Daniel Hsu

Multi-group agnostic learning is a formal learning criterion that is concernedwith the conditional risks of predictors within subgroups of a population. The criterion addresses recent practical concerns such as subgroup fairness andhidden stratification. This paper studies the structure of solutions to the multi-group learning problem, and provides simple and near-optimal algorithms for the learningproblem.

Author Information

Christopher Tosh (Memorial Sloan Kettering)
Christopher Tosh

I am a postdoc in the Department of Epidemiology and Biostatistics at Memorial Sloan Kettering Cancer Center. My current interests are on problems arising in interactive learning, representation learning, and robust learning, particularly with applications to cancer research. Previously, I was at Columbia University, where I worked with Daniel Hsu; and before that, I received my PhD from UC San Diego, where I was advised by Sanjoy Dasgupta.

Daniel Hsu (Columbia University)

More from the Same Authors