NeurIPS 2019
Sun Dec 8th through Sat the 14th, 2019 at Vancouver Convention Center
Paper ID:5730
Title:Optimizing Generalized Rate Metrics with Three Players


		
The paper was very well received. Key strengths of the paper were its analysis of the three player game involving model parameters, Lagrange variables and slack variables and how it answered open problems from earlier work in the area. The core idea of "solve learning problems with functions of classification rates using a game" was appreciated. The response addressed some lingering questions about the relation to prior work (another success of the author response process!)