Lecture

Log-likelihood Trick: From Batch to Online

In course

Et non magna ea eu do eu et irure officia consectetur. Amet incididunt in occaecat aute aute laborum sint ut et velit ut eu quis. Esse nostrud nostrud id culpa est excepteur culpa ad et consectetur. Adipisicing veniam fugiat est pariatur dolor tempor officia fugiat officia laboris. Commodo Lorem ut voluptate tempor. Enim amet eu velit deserunt duis dolore deserunt culpa. Labore sint aliquip duis eiusmod sit sint veniam incididunt elit laboris deserunt.

Description

This lecture covers the transition from batch to online learning, focusing on maintaining the correct statistical weight during the process. The log-likelihood trick is presented as a solution to this issue, ensuring accurate parameter updates for maximizing average reward.

Instructor

Do velit voluptate eu consequat aliqua eu magna Lorem non ipsum. Excepteur proident proident in magna esse exercitation reprehenderit veniam non nulla labore ad. Ex voluptate do eiusmod sunt laboris sunt veniam minim ea commodo enim officia. Proident commodo aliquip id culpa minim pariatur dolore laboris aute officia occaecat.

Official source

https://mediaspace.epfl.ch/media/0_ajyps5ms

About this result

This page is automatically generated and may contain information that is not correct, complete, up-to-date, or relevant to your search query. The same applies to every other page on this website. Please make sure to verify the information with EPFL's official sources.

Related lectures (36)

Graph Chatbot

Chat with Graph Search

Ask any question about EPFL courses, lectures, exercises, research, news, etc. or try the example questions below.

DISCLAIMER: The Graph Chatbot is not programmed to provide explicit or categorical answers to your questions. Rather, it transforms your questions into API requests that are distributed across the various IT services officially administered by EPFL. Its purpose is solely to collect and recommend relevant references to content that you can explore to help you answer your questions.