光華講壇——社會(huì)名流與企業(yè)家論壇第6588期

主題：Introduction to Multi-armed Bandits 多臂老虎機(jī)導(dǎo)論（系列講座）

主講人：伊利諾伊大學(xué)芝加哥分校 周文心副教授

主持人：西南財(cái)經(jīng)大學(xué)統(tǒng)計(jì)學(xué)院 常晉源教授

時(shí) 間：6月26日 14:00-17:00  6月27日 14:00-17:00  7月2日 14:00-17:00  7月3日 14:00-17:00

舉辦地點(diǎn)：西南財(cái)經(jīng)大學(xué)光華校區(qū)光華裙樓2303教室

主辦單位：數(shù)據(jù)科學(xué)與商業(yè)智能聯(lián)合實(shí)驗(yàn)室 統(tǒng)計(jì)學(xué)院 科研處

主講人簡介：

Wen-Xin Zhou is an Associate Professor in the Department of Information and Decision Sciences at the College of Business Administration, University of Illinois at Chicago. From 2017 to 2023, he was a faculty member in the Department of Mathematics at the University of California, San Diego. His research interests include high-dimensional statistics, robust learning for heavy-tailed data, nonparametric statistics, neural networks and deep learning, quantile regression methods and beyond.

周文心，伊利諾伊大學(xué)芝加哥分校工商管理學(xué)院信息與決策科學(xué)系副教授。2017 年至 2023 年，他在加州大學(xué)圣地亞哥分校數(shù)學(xué)系任教。他的研究興趣包括高維統(tǒng)計(jì)、重尾數(shù)據(jù)的穩(wěn)健學(xué)習(xí)、非參數(shù)統(tǒng)計(jì)、神經(jīng)網(wǎng)絡(luò)和深度學(xué)習(xí)、量化回歸方法等。

內(nèi)容簡介：

Multi-armed bandits are a simple but powerful framework for algorithms that make decisions over time under uncertainty. An enormous body of work has accumulated over the years, covered in several books and surveys. In these talks, we provide an introductory treatment of the subject. Each chapter tackles a particular line of work, offering a self-contained, teachable technical introduction and a brief review of further developments.

To begin with, we discuss a spectrum of decision-making problems, online learning, and prediction. Then, we describe some of the fundamental algorithms for multi-armed bandits. The second chapter concerns contextual bandits, focusing on stochastic linear bandits. In chapter three, we consider the stochastic sparse linear bandit problem, where only a sparse subset of context features affects the expected reward function. We will review some representative and very recent works in this direction and discuss some open questions.

多臂老虎機(jī)是一個(gè)簡單但強(qiáng)大的算法框架，用于在不確定性下隨時(shí)間做出決策。多年來，關(guān)于這一主題的大量研究成果積累了下來，并已在多本書籍和綜述中有所涵蓋。在本系列講座中，我們將對(duì)該主題進(jìn)行介紹。每一章都會(huì)討論一個(gè)特定的研究方向，提供一個(gè)自成體系的技術(shù)介紹，并簡要回顧相關(guān)的最新進(jìn)展。

首先，將討論一系列決策問題、在線學(xué)習(xí)和預(yù)測(cè)。然后，將講解一些多臂老虎機(jī)的基本算法。第二章將關(guān)注上下文老虎機(jī)，特別是隨機(jī)線性老虎機(jī)。第三章，將講解隨機(jī)稀疏線性老虎機(jī)問題，其中只有稀疏的上下文特征子集會(huì)影響期望獎(jiǎng)勵(lì)函數(shù)。最后，將回顧該方向上一些具有代表性和最新的研究工作，并討論一些未解決的問題。