Learning theory: stability is sufficient for generalization and necessary and sufficient for consistency of empirical risk minimization.

Sayan Mukherjee Partha Niyogi Tomaso Poggio Ryan M. Rifkin Learning theory: stability is sufficient for generalization and necessary and sufficient for consistency of empirical risk minimization. 161-193 2006 25 Adv. Comput. Math. 1-3 http://dx.doi.org/10.1007/s10444-004-7634-z db/journals/adcm/adcm25.html#MukherjeeNPR06