光華講壇——社會(huì)名流與企業(yè)家論壇第6651期
主題：強(qiáng)化學(xué)習(xí)基礎(chǔ)
主講人：倫敦政治經(jīng)濟(jì)學(xué)院 史成春副教授
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：11月4日09:00-12:00
舉辦地點(diǎn)：西南財(cái)經(jīng)大學(xué)光華校區(qū)光華樓1003會(huì)議室
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(jiǎn)介：
Chengchun Shi is an Associate Professor at London School of Eco- nomics and Political Science. He is serving as the associate editors of JRSSB, JASA (TM), JASA (CS) and Journal of Nonparametric Statistics. His research focuses on developing statistical learning methods in reinforcement learning, with applications to healthcare, ridesharing, video-sharing and neuroimaging. He was the recipient of the Royal Statistical Society Research Prize in 2021 and IMS Tweedie Award in 2024
史成春是倫敦經(jīng)濟(jì)學(xué)院和政治科學(xué)學(xué)院的副教授。他目前擔(dān)任《皇家統(tǒng)計(jì)學(xué)會(huì)B期刊》（JRSSB）、《美國(guó)統(tǒng)計(jì)協(xié)會(huì)期刊》（JASA，技術(shù)與方法版）、《美國(guó)統(tǒng)計(jì)協(xié)會(huì)期刊》（JASA，計(jì)算科學(xué)版）和《非參數(shù)統(tǒng)計(jì)雜志》的副主編。他的研究重點(diǎn)是開(kāi)發(fā)強(qiáng)化學(xué)習(xí)中的統(tǒng)計(jì)學(xué)習(xí)方法，并將其應(yīng)用于醫(yī)療保健、拼車(chē)、視頻分享和神經(jīng)成像等領(lǐng)域。他曾于2021年獲得皇家統(tǒng)計(jì)學(xué)會(huì)研究獎(jiǎng)，并在2024年獲得了IMS Tweedie獎(jiǎng)。
內(nèi)容簡(jiǎn)介：
Reinforcement learning (RL, see Sutton and Barto, 2018, for an overview) is a powerful machine learning technique that allows an agent to learn and interact with a given environment, to maximize the cumulative reward the agent receives. It has been one of the most popular research topics in the machine learning and computer science literature over the past few years. Significant progress has been made in solving challenging problems across various domains using RL, including games, recommender systems, finance, healthcare, robotics, transportation. This lecture mainly focusses on foundations of Reinforcement Learning. We will also provide code to implement various RL algorithms discussed in the lecture. The materials of this course are available on https://github.com/callmespring/RL-short-course.
強(qiáng)化學(xué)習(xí)（RL，見(jiàn) Sutton 和 Barto，2018的概述）是一種強(qiáng)大的機(jī)器學(xué)習(xí)技術(shù)，它允許一個(gè)代理學(xué)習(xí)并與給定的環(huán)境互動(dòng)，以最大化代理收到的累積獎(jiǎng)勵(lì)。在過(guò)去幾年中，它一直是機(jī)器學(xué)習(xí)和計(jì)算機(jī)科學(xué)文獻(xiàn)中最流行的研究主題之一。在各種領(lǐng)域使用 RL 解決挑戰(zhàn)性問(wèn)題方面取得了顯著進(jìn)展，包括游戲、推薦系統(tǒng)、金融、醫(yī)療保健、機(jī)器人技術(shù)和交通。此次講座主要介紹強(qiáng)化學(xué)習(xí)基礎(chǔ)。我們還將提供代碼來(lái)實(shí)現(xiàn)講座中討論的各種 RL 算法。這個(gè)課程的材料可以在https://github.com/callmespring/RL-short-course上找到。
