國立虎尾科技大學 |

登入

回首頁

The alignment problem : = how can artificial intelligence learn human values? /

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	The alignment problem :/ Brian Christian.
其他題名:	how can artificial intelligence learn human values? /
作者:	Christian, Brian,
出版者:	London :Atlantic Books, : 2021.,
面頁冊數:	x, 476 p. ;20 cm.;
附註:	"First published in the United States in 2020 by W.W. Norton & Company, Inc., New York".
標題:	Social values. -
ISBN:	9781786494337 (pbk.) :

The alignment problem : = how can artificial intelligence learn human values? /
Christian, Brian,1984-

The alignment problem :how can artificial intelligence learn human values? /Brian Christian. - Pbk. ed. - London :Atlantic Books,2021. - x, 476 p. ;20 cm.

"First published in the United States in 2020 by W.W. Norton & Company, Inc., New York".

Includes bibliographical references (p. [401]-451) and index.

I. Prophecy -- 1. Representation -- 2. Fairness -- 3. Transparency -- II. Agency -- 4. Reinforcement -- 5. Shaping -- 6. Curiosity -- III. Normativity -- 7. Imitation -- 8. Inference -- 9. Uncertainty.

"How do we prevent AI working against us? Artificial intelligence is rapidly dominating every aspect of our modern lives influencing the news we consume, whether we get a mortgage, and even which friends wish us happy birthday. But as algorithms make ever more decisions on our behalf, how do we ensure they do what we want? And fairly? This conundrum - dubbed 'The Control Problem' by experts - is the subject of this timely and important book. From the AI program which cheats at computer games to the sexist algorithm behind Google Translate, bestselling author Brian Christian explains how, as AI develops, we rapidly approach a collision between artificial intelligence and ethics. If we stand by, we face a future with unregulated algorithms that propagate our biases - and worse - violate our most sacred values. Urgent and fascinating, this is an accessible primer to the most important issue facing Al researchers today"--Publisher's description.

ISBN: 9781786494337 (pbk.) :NT531Subjects--Topical Terms:

555260
Social values.

LC Class. No.: Q334.7 / .C47 2021

Dewey Class. No.: 174/.90063

The alignment problem : = how can artificial intelligence learn human values? /
LDR:02048cam a2200253 a 4500 001 1158479
005 20250827090916.0
008 250918s2021 enk b 001 0 eng d
020 $a 9781786494337 (pbk.) : $c NT531
020 $a 1786494337 (pbk.)
020 $a 9781786494320 (ebk.)
035 $a (OCoLC)1287954563
035 $a on1287954563
040 $a AU@ $b eng $c AU@ $d OCLCF $d OCLCO $d UKOBU $d CDX $d NZAUC $d OCLCQ $d OCLCL $d OCLCQ $d OCLCL $d NFU
041 0 # $a eng
050 # 4 $a Q334.7 $b .C47 2021
082 0 4 $a 174/.90063 $2 23
100 1 $a Christian, Brian, $d 1984- $3 1484960
245 1 4 $a The alignment problem : $b how can artificial intelligence learn human values? / $c Brian Christian.
250 $a Pbk. ed.
260 # $a London : $b Atlantic Books, $c 2021.
300 $a x, 476 p. ; $c 20 cm.
500 $a "First published in the United States in 2020 by W.W. Norton & Company, Inc., New York".
504 $a Includes bibliographical references (p. [401]-451) and index.
505 0 # $a I. Prophecy -- 1. Representation -- 2. Fairness -- 3. Transparency -- II. Agency -- 4. Reinforcement -- 5. Shaping -- 6. Curiosity -- III. Normativity -- 7. Imitation -- 8. Inference -- 9. Uncertainty.
520 # $a "How do we prevent AI working against us? Artificial intelligence is rapidly dominating every aspect of our modern lives influencing the news we consume, whether we get a mortgage, and even which friends wish us happy birthday. But as algorithms make ever more decisions on our behalf, how do we ensure they do what we want? And fairly? This conundrum - dubbed 'The Control Problem' by experts - is the subject of this timely and important book. From the AI program which cheats at computer games to the sexist algorithm behind Google Translate, bestselling author Brian Christian explains how, as AI develops, we rapidly approach a collision between artificial intelligence and ethics. If we stand by, we face a future with unregulated algorithms that propagate our biases - and worse - violate our most sacred values. Urgent and fascinating, this is an accessible primer to the most important issue facing Al researchers today"--Publisher's description.
650 # 0 $a Social values. $3 555260
650 # 0 $a Software failures. $3 1028116
650 # 0 $a Machine learning $x Safety measures. $3 1484961
650 # 0 $a Artificial intelligence $x Social aspects. $3 562911
650 # 0 $a Artificial intelligence $x Moral and ethical aspects. $3 796961
650 # 0 $a Ethical problems. $3 643565