Title :
Improved Risk Tail Bounds for On-Line Algorithms
Author :
Cesa-Bianchi, Nicolò ; Gentile, Claudio
Author_Institution :
Milano Univ., Milano
Abstract :
Tight bounds are derived on the risk of models in the ensemble generated by incremental training of an arbitrary learning algorithm. The result is based on proof techniques that are remarkably different from the standard risk analysis based on uniform convergence arguments, and improves on previous bounds published by the same authors.
Keywords :
learning (artificial intelligence); risk analysis; statistical analysis; arbitrary learning algorithm; ensemble; incremental training; online algorithms; proof techniques; risk analysis; risk tail bounds; statistical learning theory; uniform convergence arguments; Algorithm design and analysis; Convergence; Inference algorithms; Information processing; Loss measurement; Performance loss; Risk analysis; Standards publication; Statistical learning; Tail; Martingales; on-line learning; risk bounds; statistical learning theory;
Journal_Title :
Information Theory, IEEE Transactions on
DOI :
10.1109/TIT.2007.911292
