Loss Functions [chapter]

Robert C. Williamson
2013 Empirical Inference  
Vapnik described the "three main learning problems" of pattern recognition, regression estimation and density estimation. These are defined in terms of the loss functions used to evaluate performance (0-1 loss, squared loss and log loss respectively). But there are many other loss functions one could use. In this chapter I will summarise some recent work by myself and colleagues studying the theoretical aspects of loss functions. The results elucidate the richness of the set of loss functions
more » ... d explain some of the implications of their choice.
doi:10.1007/978-3-642-41136-6_8 dblp:conf/birthday/Williamson13 fatcat:ws5ry6sf3jbu5eoj7vcgbgelsq