...

from Wikipedia

by user

on
Category: Documents
55

views

Report

Comments

Transcript

from Wikipedia
Lecture 1 of Artificial Intelligence
Introduction to
Artificial Intelligence
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/1
Topics of this lecture
•
•
•
•
What is artificial intelligence?
Related research fields
A brief review of AI history
Some key persons
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/2
知能とは?
• Wikipedia
– 知能とは、論理的に考える、計画を立てる、問題を解決す
る、抽象的に考える、考えを把握する、言語機能、学習機
能などさまざまな知的活動を含む心の特性のことである。
• 広辞苑
– 知識と才能;知性の程度;
– 〔心〕環境に適応し、新しい問題状況に対処する知的機能、
能力。
• 国語・新語辞書
– 知識と才能、知恵のはたらき;
– 〔心〕 学習し、抽象的な思考をし、環境に適応する知的機
能のもとになっている能力。
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/3
What is intelligence?
• Intelligence (also called intellect) is an umbrella
term used to describe a property of the mind that
encompasses many related abilities, such as the
capacities to reason, to plan, to solve problems,
to think abstractly, to comprehend ideas, to use
language, and to learn (from Wikipedia)
Where is my mind?
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/4
知と智
Intelligence and Wisdom
• 「知」の意味
– その存在を認めている。認識
する。
「事件の発生を知る」「昔から
知っていたことだ」
– それについての知識を有する。
わきまえる。
「知らない土地で知った人に会
う」
– その内容・意味などを理解す
る。悟る。
「一を聞いて十を知る」
• 「智」の意味
– 物の道理を知り、正しい判断
を下す能力。儒教における五
常の一(仁、義、理、智、信)。
– (ア)「知識」に同じ。
(イ)知識を獲得するはたらき
– 〔仏〕〔梵 j na〕慧(え)の一。
真理に従って判断し、煩悩
(ぼんのう)を打ち消す精神の
はたらき。《智》
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
大智若愚
AI Lec01/5
知と智
Intelligence and Wisdom
• Intelligenceは日本語では「知能」と訳す、中国語で
は「智能」と訳す。
• Wisdomは日本語では「知恵」となり、中国語では
「智慧」となる(日本語の「智慧」は宗教関係者しか
使わないらしい)。
• 日本人は「知るチカラ」(知識)を、中国人は「知るコ
コロ」(智慧)をそれぞれ重視していると考えられる。
• 実際、両方をバランスよく取らないと難しい問題の
解決はできない。
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/6
What is AI ?
• Textbooks often define
artificial intelligence as
“the study and design of
computing systems that
perceives its environment
and takes actions like
human beings”.
• The term was introduced
by John McCarthy in
1956 in the well-known
Dartmouth Conference.
• 人工知能とは、通常の教科
書では、「コンピュータに人
間と同様の知能を実現させ
ようという試み、あるいはそ
のための一連の基礎技術
をさす」と定義する。
• 「人工知能」という名前は
1956年にダートマス会議で
ジョン・マッカーシーにより
命名された。
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/7
Different definitions of AI
(from “Artificial Intelligence: A Modern Approach”)
Systems that think like
humans
人間のように
考える機械
Systems that think
rationally
正しいことを
考える機械
Systems that act like
humans
人間のように
行う機械
Systems that act
rationally
正しいことを
行う機械
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/8
Related research fields
Search
Knowledge representation
Reasoning
Learning and understanding
Problem solving
Planning
Recognition
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
探索
知識表現
推論
学習と理解
問題解決
企画
認識
AI Lec01/9
Search (探索)
• Brute-force search
– Depth-first search
– Breadth-first search
• Heuristic Search
– Hill climbing search
– Best-first search
– A* Algorithm
• 網羅的探索
– 縦型(深さ優先)探索
– 横型(幅優先)探索
• 発見的探索
– 山登り法
– 最良優先探索
– A* アルゴリズム
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/10
知識獲得の3要素
• What is the input?
– Map from real world to the mind model
• What is the output?
– Map from the mind model to the real world
• What is the relation between the input
and the output?
– Abstraction of the real world
現実世界
問題表現
学習モデル
と方法
知識表現
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
現実世界
との対応
AI Lec01/11
Representation methods
表現方法
• Representation of the
problem
• 問題表現
– 状態空間表現
– State space representation
• Representation of
knowledge
–
–
–
–
–
Production system
Semantic network
Predicate logic
Fuzzy logic
Neural network
• 知識表現
– プロダクションシス
テム
– 意味ネット
– 述語論理
– ファジー論理
– 神経回路網
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/12
学習モデルとアルゴリズム
•
•
•
•
Neural network learning
Evolutionary learning
Reinforcement learning
Artificial immune
system
• Fuzzy logic
• Decision tree
• Hybrid system
•
•
•
•
•
•
•
ニューラルネット
進化学習
強化学習
人工免疫システム
ファジー論理
決定木
ハイブリッドシステム
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/13
AIの歴史:播種期(20世紀はじめ頃)
• 代表人物
– George Boole
– Alfred North Whitehead
– Bertrand A. W. Russell
• 主な成果
– ブール代数
– 数学原理 - 論理主義
– ラッセルのパラドックス
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/14
George Boole (1815-1864)
• George Boole was not a
computer scientist.
• Boolean algebra was
developed by him.
• This has become one of the
mathematic foundations of
computer science.
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
(from Wikipedia)
AI Lec01/15
Alfred North Whitehead (1861- 1947)
• Alfred North Whitehead was
an English mathematician
who became a philosopher.
• He wrote on algebra, logic,
foundations of mathematics,
philosophy of science, physics,
metaphysics, and education.
• He co-authored the epochal
Principia Mathematica with
Bertrand Russell.
(from Wikipedia)
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/16
Bertrand Arthur William Russell, 3rd Earl
Russell (1872–1970)
• Bertrand Arthur William Russell,
3rd Earl Russell was a philosopher,
historian, logician, mathematician,
advocate for social reform, and pacifist.
• A prolific writer, he was a popularizer
of philosophy and a commentator on a
large variety of topics.
• He was a prominent anti-war activist,
championing free trade between
nations and anti-imperialism.
• He wrote the essay On Denoting and
was co-author (with Alfred North
Whitehead) of Principia Mathematica,
an attempt to ground mathematics on
the laws of logic.
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
(from Wikipedia)
AI Lec01/17
AIの歴史:萌芽期(20世紀前半)
• 代表人物
– Alan Turing
– Claude Shannon
– John von Neumann
• 主な成果
– 計算理論、Turing Machine, Turing test
– 情報理論、ブール代数に基づく論理
– 計算機の基本構成、von Neumann machine
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/18
Alan Turing (1912-1954)
• Turing is often considered to
be the father of modern
computer science.
• Turing provided an influential
formalization of the concept of
the algorithm and computation
with the Turing machine.
• With the Turing test, he made
a significant and
characteristically provocative
contribution to the debate
regarding artificial intelligence:
whether it will ever be possible
to say that a machine is
conscious and can think.
(from Wikipedia)
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/19
Claude Shannon(1916-2001)
• Shannon, an American electrical
engineer and mathematician, was
"the father of information theory“.
• He is also credited with founding
both digital computer and digital
circuit design theory in 1937, when,
as a 21-year-old master's student
at MIT, he wrote a thesis
demonstrating that electrical
application of Boolean algebra
could construct and resolve any
logical, numerical relationship.
(from Wikipedia)
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/20
John von Neumann (1903-1957)
• John von Neumann was a
Hungarian-American mathematician
who made major contributions to a
vast range of fields including
–
–
–
–
–
–
set theory
functional analysis
quantum mechanics
ergodic theory
economics and game theory
computer science
• The so called conventional CPU
based computer was proposed by him,
and he is generally regarded as one of
the foremost mathematicians of the
20th century.
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
(from Wikipedia)
AI Lec01/21
成長期(20世紀60年代)
• 代表人物
– John McCarthy
– Marvin Lee Minsky
– Herbert Alexander Simon
– Allen Newell
• 主な成果
– General problem solver
– LISP
– Expert system
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/22
John McCarthy
• John McCarthy (born
September 4, 1927, in Boston,
Massachusetts), is an
American computer scientist
and cognitive scientist.
• He received the Turing Award
in 1971 for his major
contributions to the field of
Artificial Intelligence (AI).
• He was responsible for the
coining of the term "Artificial
Intelligence" in his 1955
proposal for the 1956
Dartmouth Conference and is
the inventor of the Lisp
programming language.
(from Wikipedia)
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/23
Marvin Lee Minsky
• Marvin Lee Minsky (born
August 9, 1927) is an
American cognitive scientist in
the field of artificial intelligence
(AI), co-founder of MIT's AI
laboratory, and author of
several texts on AI and
philosophy.
• Minsky won the Turing Award
in 1969, the Japan Prize in
1990, the IJCAI Award for
Research Excellence in 1991,
and the Benjamin Franklin
Medal from the Franklin
Institute in 2001.
(from Wikipedia)
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/24
Herbert Alexander Simon (1916-2001)
• Herbert Alexander Simon was an American
political scientist whose research ranged across
the fields of cognitive psychology, computer
science, public administration, etc.
• Simon was a truly innovative thinker. He was
among the founding fathers of several of today's
most important scientific domains, including
artificial Intelligence, information processing,
decision-making, problem-solving, etc.
• He coined the terms bounded rationality and
satisficing, and was the first to analyze the and
to propose a preferential attachment mechanism
to explain power law distributions.
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
(from Wikipedia)
AI Lec01/25
Allen Newell (1927-1992)
• Allen Newell was a researcher in
computer science and cognitive
psychology at the RAND corporation
and at Carnegie Mellon University’s
School of Computer Science.
• He contributed to the Information
Processing Language (1956) and two
of the earliest AI programs, the Logic
Theory Machine (1956) and the
General Problem Solver (1957) (with
Herbert Simon).
• He was awarded the ACM's A.M.
Turing Award along with Herbert Simon
in 1975 for their basic contributions to
artificial intelligence and the psychology
of human cognition.
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
(from Wikipedia)
AI Lec01/26
Edward Albert Feigenbaum
• Edward Albert Feigenbaum (born
January 20, 1936) is a computer scientist
working in the field of artificial intelligence.
• He is often called the "father of expert
systems."
• In his PhD thesis, carried out under the
supervision of Herbert Simon, he
developed EPAM, one of the first computer
models of how people learn.
• He received the ACM Turing Award, jointly
with Raj Reddy in 1994 "For pioneering the
design and construction of large scale
artificial intelligence systems,
demonstrating the practical importance and
potential commercial impact of artificial
intelligence technology".
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
(from Wikipedia)
AI Lec01/27
人工知能の近況
• 1980年代:知識の獲得と人間らし
い推論
• ニューラルネットワーク、進化、ファジィ
の流行
• ソフトコンピューティング
• 2010年代:バーチャル空間における学習
• 統計的学習:因子解析、動向解析など
• オントロジー工学:マシン間の知識共有
• 深層学習:ニューラルネットワークの拡張
• 2016年:アルファ碁の衝撃
• 最も難しいゲームで人間のプロに勝った。
• AIがいよいよ知能の頂点に立つ。
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/28
今週の課題
• この授業で紹介した人工知能の発展に重要な貢献をした人
物の中から一人を選び、その人の業績を小作文の形で詳し
く紹介せよ。
– 作文は、800文字程度、A4サイズ用紙一枚、日本語と英語のどちら
で書かれてもよい。
– 資料の収集は、インターネットで行えるが、参考した文献の情報(著
者、出版社、時期、論文ならそのタイトル、ウエブサイトならそのURL
など)を文書の最後に明記すること。
– 著作権関連の法律を守り、資料から丸ごと写すことを禁止する。
– 文書の編集は、どのツールを使ってもよいが、最後は必ずpdf形式に
変換し、指定した場所に置くこと。
– 作文のタイトルの下に、自分の学籍番号と氏名を記入すること。
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/29
課題について
•
•
•
ホームディレクトリにAIディレクトリを作成する.
AIディレクトリ内にex_XXというディレクトリを作って作業を行う.
TAがこれを回収できるようにパーミッションを705にすること.
–
–
•
ex_01というサブディレクトリを作成し、その下に解答ファイルanswer.pdfを入れる。
ex_01とanswer.pdfのパーミッションを705にすること。
第2回以降の課題は演習ごとにzip形式で授業ホームページにアップロードされている.
–
–
•
mkdir ~/AI
chmod 705 ~/AI
第1回の課題については、
–
–
•
提出期限: 7日(一週間)後、授業クラス
が始まる前 (タイムスタンプで判断する)
http://web-ext.u-aizu.ac.jp/~qf-zhao/TEACHING/AI/AI.html
ファイル名はex_XXという形式になっている. (XX = 02, 03,…, 13)
アーカイブ内のファイルについて
–
–
–
–
–
–
–
README.md
*.c, *.h
Makefile
LICENSE
data_XX.txt
answer_XX.txt
summary_XX.txt
:
:
:
:
:
:
:
課題内容についての説明
ソースファイル、演習ではこれを編集する.
makeスクリプト (後述)
ソースファイルのライセンス情報
プログラムの入力データ
正しい出力
変更点の説明、及び考察が必要な場合はそれを書くこと.
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/30
例: ex_02
•
課題をダウンロードし、AIディレクトリ内に展開、作業を行う.
–
–
–
•
課題をダウンロードするたびに、パーミッションを705にすること.
–
•
•
chmod -R 705 ./ex_02/
README.mdを参照し、ソースファイルを修正する.
makeコマンドでプログラムを実行する.
–
–
–
–
•
•
cd ~/AI
wget http://www.u-aizu.ac.jp/~qf-zhao/TEACHING/AI/ex_02.zip
unzip ex_02.zip
make test
> gcc -Wall -std=c99 -O2 -o ./a.out prog_02.c func.c queue.c
> ./a.out < data_02.txt > result_02.txt
> diff -y --suppress-common-lines answer_02.txt result_02.txt
make testで回答と出力のdiffを取り、正しいかを確認している.
上記のようにdiffの出力が無くなるようにプログラムを修正していくこと.
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/31
今週のクイズ
(授業の中で回答し、授業終了時に提出すること)
•
•
AIの4つの定義はなんですか?
探索はなんですか?
•
知識獲得の3要素はなんですか?
•
学習モデルの例を2つ挙げてください。
•
AIの歴史上、各時期の代表人物を一人ずつ挙げてくだ
さい。
Produced by Qiangfu Zhao (Since 2008), All rights reserved ©
AI Lec01/32
Fly UP