From Proper Scoring Rules to Max-Min Optimal Forecast Aggregation

There are many ways to elicit honest probabilistic forecasts from experts. Once those elicited, there aggregate them into a single forecast. Should the choice of elicitation method inform aggregation method? In “From Proper Scoring Rules Max-Min Optimal Forecast Aggregation,” Neyman and Roughgarden establish connection between these two problems. To every they associate that improves as much possible upon forecast randomly chosen expert, in worst case. This association maps most widely used methods (Brier logarithmic scoring) well-known (linear pooling). The authors show number interesting properties this connection, including natural axiomatization obtained through well an algorithm for efficient no-regret learning expert weights.