蒙特卡羅方法與人工智能

定　價(jià)：￥138.00

作　者：	（美）Adrian Barbu（巴布·艾?。琒ong-Chun Zhu（朱松純）
出版社：	電子工業(yè)出版社
叢編項(xiàng)：
標(biāo)　簽：	暫缺

購(gòu)買這本書可以去

當(dāng)當(dāng)網(wǎng) (￥103.50)

ISBN：	9787121470202	出版時(shí)間：	2024-01-01	包裝：	平裝-膠訂
開(kāi)本：	16開(kāi)	頁(yè)數(shù)：		字?jǐn)?shù)：

內(nèi)容簡(jiǎn)介

　　本書全面敘述了蒙特卡羅方法，包括序貫蒙特卡羅方法、馬爾可夫鏈蒙特卡羅方法基礎(chǔ)、Metropolis算法及其變體、吉布斯采樣器及其變體、聚類采樣方法、馬爾可夫鏈蒙特卡羅的收斂性分析、數(shù)據(jù)驅(qū)動(dòng)的馬爾可夫鏈蒙特卡羅方法、哈密頓和朗之萬(wàn)蒙特卡羅方法、隨機(jī)梯度學(xué)習(xí)和可視化能級(jí)圖等。為了便于學(xué)習(xí)，每章都包含了不同領(lǐng)域的代表性應(yīng)用實(shí)例。本書旨在統(tǒng)計(jì)學(xué)和計(jì)算機(jī)科學(xué)之間架起一座橋梁以彌合它們之間的鴻溝，以便將其應(yīng)用于計(jì)算機(jī)視覺(jué)、計(jì)算機(jī)圖形學(xué)、機(jī)器學(xué)習(xí)、機(jī)器人學(xué)、人工智能等領(lǐng)域解決更廣泛的問(wèn)題，同時(shí)使這些領(lǐng)域的科學(xué)家和工程師們更容易地利用蒙特卡羅方法加強(qiáng)他們的研究。本書適合計(jì)算機(jī)、人工智能、機(jī)器人等領(lǐng)域的教師、學(xué)生閱讀和參考，也適合相關(guān)領(lǐng)域的研究者和工業(yè)界的從業(yè)者閱讀。

作者簡(jiǎn)介

　　朱松純，1996年獲得哈佛大學(xué)計(jì)算機(jī)科學(xué)博士學(xué)位，現(xiàn)任北京通用人工智能研究院院長(zhǎng)、北京大學(xué)人工智能研究院院長(zhǎng)、北京大學(xué)講席教授、清華大學(xué)基礎(chǔ)科學(xué)講席教授；曾任美國(guó)加州大學(xué)洛杉磯分校（UCLA）統(tǒng)計(jì)學(xué)與計(jì)算機(jī)科學(xué)教授，加州大學(xué)洛杉磯分校視覺(jué)、認(rèn)知、學(xué)習(xí)與自主機(jī)器人中心主任。他長(zhǎng)期致力于為視覺(jué)和智能探尋一個(gè)統(tǒng)一的統(tǒng)計(jì)與計(jì)算框架：包括作為學(xué)習(xí)與推理的統(tǒng)一表達(dá)和數(shù)字蒙特卡羅方法的時(shí)空因果與或圖（STC-AOG）。他在計(jì)算機(jī)視覺(jué)、統(tǒng)計(jì)學(xué)習(xí)、認(rèn)知、人工智能和自主機(jī)器人領(lǐng)域發(fā)表了400多篇學(xué)術(shù)論文。他曾獲得了多項(xiàng)榮譽(yù)，2003年因圖像解析的工作成就獲馬爾獎(jiǎng)，1999年因紋理建模、2007年因物體建模兩次獲得馬爾獎(jiǎng)提名。2001 年，他獲得了NSF青年科學(xué)家獎(jiǎng)、ONR青年研究員獎(jiǎng)和斯隆獎(jiǎng)。因?yàn)樵谝曈X(jué)模式的概念化、建模、學(xué)習(xí)和推理的統(tǒng)一基礎(chǔ)方面的貢獻(xiàn)，他2008年獲得了國(guó)際模式識(shí)別協(xié)會(huì)授予的J.K. Aggarwal獎(jiǎng)。2013 年，他關(guān)于圖像分割的論文獲得了亥姆霍茲獎(jiǎng)（Helmholtz Test-of-Time Award）。2017年，他因生命度建模工作獲國(guó)際認(rèn)知學(xué)會(huì)計(jì)算建模獎(jiǎng)。2011年，他當(dāng)選IEEE Fellow。他兩次擔(dān)任國(guó)際計(jì)算機(jī)視覺(jué)與模式識(shí)別大會(huì)（CVPR 2012，2019）主席。作為項(xiàng)目負(fù)責(zé)人，他領(lǐng)導(dǎo)了多個(gè)ONR MURI和DARPA團(tuán)隊(duì)，從事統(tǒng)一數(shù)學(xué)框架下的場(chǎng)景和事件理解以及認(rèn)知機(jī)器人的工作。巴布·艾俊，2000 年獲得俄亥俄州立大學(xué)數(shù)學(xué)博士學(xué)位，2005 年獲得加州大學(xué)洛杉磯分校計(jì)算機(jī)科學(xué)博士學(xué)位（師從朱松純博士）。2005年至2007年，他在西門子研究院從事醫(yī)學(xué)成像研究工作，從開(kāi)始擔(dān)任研究科學(xué)家到后來(lái)升任項(xiàng)目經(jīng)理。由于在邊緣空間學(xué)習(xí)方面的工作成就，他與西門子的合作者獲得了2011年Thomas A. Edison專利獎(jiǎng)。2007年，他加入佛羅里達(dá)州立大學(xué)統(tǒng)計(jì)系，從助理教授到副教授，再到2019年擔(dān)任教授。他發(fā)表了70多篇關(guān)于計(jì)算機(jī)視覺(jué)、機(jī)器學(xué)習(xí)和醫(yī)學(xué)成像方面的論文，并擁有超過(guò)25項(xiàng)與醫(yī)學(xué)成像和圖像去噪相關(guān)的專利。魏平，西安交通大學(xué)人工智能學(xué)院教授、博士生導(dǎo)師，人工智能學(xué)院副院長(zhǎng)，國(guó)家級(jí)青年人才，陜西高校青年創(chuàng)新團(tuán)隊(duì)（自主智能系統(tǒng)）帶頭人，西安交通大學(xué)“青年拔尖人才支持計(jì)劃”A類入選者。西安交通大學(xué)學(xué)士、博士學(xué)位，美國(guó)加州大學(xué)洛杉磯分校（UCLA）博士后、聯(lián)合培養(yǎng)博士。研究領(lǐng)域包括計(jì)算機(jī)視覺(jué)、機(jī)器學(xué)習(xí)、智能系統(tǒng)等。主持國(guó)家自然科學(xué)基金項(xiàng)目、國(guó)家重點(diǎn)研發(fā)計(jì)劃子課題等科研項(xiàng)目十余項(xiàng)，作為骨干成員參與國(guó)家自然科學(xué)基金重大科學(xué)研究計(jì)劃等課題多項(xiàng)。在TPAMI、CVPR、ICCV、ACM MM、AAAI、IJCAI等國(guó)際權(quán)威期刊和會(huì)議發(fā)表學(xué)術(shù)論文多篇，是十余個(gè)國(guó)際著名期刊和會(huì)議審稿人。擔(dān)任中國(guó)自動(dòng)化學(xué)會(huì)網(wǎng)聯(lián)智能專委會(huì)副主任委員、中國(guó)圖象圖形學(xué)學(xué)會(huì)機(jī)器視覺(jué)專委會(huì)委員。

圖書目錄

目錄
第1 章蒙特卡羅方法簡(jiǎn)介··············································································.1
1.1 引言·······························································································.1
1.2 動(dòng)機(jī)和目標(biāo)······················································································.1
1.3 蒙特卡羅計(jì)算中的任務(wù)·······································································.2
1.3.1 任務(wù)1：采樣和模擬········································································.3
1.3.2 任務(wù)2：通過(guò)蒙特卡羅模擬估算未知量···················································.5
1.3.3 任務(wù)3：優(yōu)化和貝葉斯推理································································.7
1.3.4 任務(wù)4：學(xué)習(xí)和模型估計(jì)···································································.8
1.3.5 任務(wù)5：可視化能級(jí)圖·····································································.9
本章參考文獻(xiàn)··························································································13
第2 章序貫蒙特卡羅方法··············································································14
2.1 引言·······························································································14
2.2 一維密度采樣···················································································14
2.3 重要性采樣和加權(quán)樣本·······································································15
2.4 序貫重要性采樣（SIS） ······································································18
2.4.1 應(yīng)用：表達(dá)聚合物生長(zhǎng)的自避游走························································18
2.4.2 應(yīng)用：目標(biāo)跟蹤的非線性/粒子濾波·······················································20
2.4.3 SMC 方法框架總結(jié)·········································································23
2.5 應(yīng)用：利用SMC 方法進(jìn)行光線追蹤·······················································24
2.6 在重要性采樣中保持樣本多樣性···························································25
2.6.1 基本方法····················································································25
2.6.2 Parzen 窗討論··············································································28
2.7 蒙特卡羅樹(shù)搜索················································································29
2.7.1 純蒙特卡羅樹(shù)搜索··········································································30
2.7.2 AlphaGo ·····················································································32
2.8 本章練習(xí)·························································································33
本章參考文獻(xiàn)··························································································35
第3 章馬爾可夫鏈蒙特卡羅方法基礎(chǔ)·······························································36
3.1 引言·······························································································36
蒙特卡羅方法與人工智能
·X ·
3.2 馬爾可夫鏈基礎(chǔ)················································································37
3.3 轉(zhuǎn)移矩陣的拓?fù)洌哼B通與周期······························································38
3.4 Perron-Frobenius 定理··········································································41
3.5 收斂性度量······················································································42
3.6 連續(xù)或異構(gòu)狀態(tài)空間中的馬爾可夫鏈·····················································44
3.7 各態(tài)遍歷性定理················································································45
3.8 通過(guò)模擬退火進(jìn)行MCMC 優(yōu)化·····························································46
3.9 本章練習(xí)·························································································49
本章參考文獻(xiàn)··························································································51
第4 章 Metropolis 算法及其變體······································································52
4.1 引言·······························································································52
4.2 Metropolis-Hastings 算法······································································52
4.2.1 原始Metropolis-Hastings 算法······························································53
4.2.2 Metropolis-Hastings 算法的另一形式·······················································54
4.2.3 其他接受概率設(shè)計(jì)··········································································55
4.2.4 Metropolis 算法設(shè)計(jì)中的關(guān)鍵問(wèn)題·······························4