The 2021 SIGIR workshop on eCommerce is hosting the Coveo Data Challenge for "In-session prediction for purchase intent and recommendations".
We first present a general risk bound for ensembles that depends on the Lp norm of the weighted combination of voters which can be selected from a continuous set.
We propose an extensive analysis of the behavior of majority votes in binary classification.
The C-bound, introduced in Lacasse et al., gives a tight upper bound on the risk of a binary majority vote classifier.
This paper generalizes an important result from the PAC-Bayesian literature for binary classification to the case of ensemble methods for structured outputs.