http://leon.bottou.org/ 2012-02-04T00:54:06-05:00 http://leon.bottou.org/ http://leon.bottou.org/lib/tpl/sidebar/images/favicon.ico text/html 2012-01-31T19:20:31-05:00 leonb http://leon.bottou.org/talks/onepass?rev=1328055631&do=diff This talk was first given in the Sixth Annual Machine Learning Symposium of the New York Academy of Sciences and in the NIPS 2011 Workshop on Computational Trade-offs in Statistical Learning. Summary The goal of the presentation is to describe practical stochastic gradient algorithms that process each training example only once, yet asymptotically match the performance of the true optimum. This statement needs, of course, to be made more precise. To achieve this, we'll review the works of Neve… text/html 2012-01-31T19:09:23-05:00 leonb http://leon.bottou.org/talks?rev=1328054963&do=diff This page contains pointers to my most significant lectures. All the slides are available under both the PDF and DjVu formats. * The BackPropagation CookBook * Graph Transformer Networks * DjVu: Scanned Documents on the Web * Online Learning and Stochastic Approximations * The Tradeoffs of Large Scale Learning * Stochastic Algorithms for One Pass Learning text/html 2011-11-23T00:51:28-05:00 leonb http://leon.bottou.org/news/vapnik-chervonenkis_sauer?rev=1322027488&do=diff Many machine learning authors write that a certain fundamental combinatorial result was independently established by Vapnik and Chervonenkis (1971), Sauer (1972), Shelah (1972), and sometimes Perles and Shelah (reference unknown). Vapnik and Chervonenkis published a version of their results in the Proceedings of the USSR Academy of Sciences four years earlier in 1968. It also appears that Sauer and Shelah pursued this result for very different purposes.