Artificial Intelligence and interdisciplinarity Bert Kappen Symposium Neuroscience Oktober 2012.

Slides:



Advertisements
Similar presentations
Artificial Intelligence
Advertisements

CSCI 4800 ARTIFICIAL INTELLIGENCE
ARTIFICIAL INTELLIGENCE
By John E. Hopcroft, Rajeev Motwani and Jeffrey D. Ullman
Additional Topics ARTIFICIAL INTELLIGENCE
Computer Literacy BASICS
Artificial Intelligence
1 Using Bayesian Network for combining classifiers Leonardo Nogueira Matos Departamento de Computação Universidade Federal de Sergipe.
Converting a Fraction to %
An Introduction to Artificial Intelligence Presented by : M. Eftekhari.
AI 授課教師:顏士淨 2013/09/12 1. Part I & Part II 2  Part I Artificial Intelligence 1 Introduction 2 Intelligent Agents Part II Problem Solving 3 Solving Problems.
A Brief History of Artificial Intelligence
Artificial Intelligence A Modern Approach Dennis Kibler.
CSCE101 –Chapter 8 (continued) Tuesday, December 5, 2006.
Introduction to Introduction to Artificial Intelligence Henry Kautz.
Overview and History of Cognitive Science
Overview and History of Cognitive Science. How do minds work? What would an answer to this question look like? What is a mind? What is intelligence? How.
Artificial Intelligence Overview John Paxton Montana State University August 14, 2003.
CSE 471/598,CBS598 Introduction to Artificial Intelligence Fall 2004
CS 480 Lec 2 Sept 4 complete the introduction Chapter 3 (search)
Random Administrivia In CMC 306 on Monday for LISP lab.
Artificial Intelligence Instructor: Professor Yun Peng
INSTRUCTOR: DR. XENIA MOUNTROUIDOU CS CS Artificial Intelligence.
ARTIFICIAL INTELLIGENCE Introduction: Chapter Textbook: S. Russell and P. Norvig Artificial Intelligence: A Modern Approach Prentice Hall, 2003,
Artificial Intelligence
ARTIFICIAL INTELLIGENCE
CPSC 171 Artificial Intelligence Read Chapter 14.
Artificial Intelligence Andrea Danyluk For FPS 2/2/04.
FOUNDATIONS OF ARTIFICIAL INTELLIGENCE Introduction: Chapter 1.
ARTIFICIAL INTELLIGENCE Introduction: Chapter 1. Outline Course overview What is AI? A brief history The state of the art.
1 AI and Agents CS 171/271 (Chapters 1 and 2) Some text and images in these slides were drawn from Russel & Norvig’s published material.
The Thinking Machine Based on Tape. Computer Has Some Intelligence Now Playing chess Solving calculus problems Other examples:
CISC4/681 Introduction to Artificial Intelligence1 Introduction – Artificial Intelligence a Modern Approach Russell and Norvig: 1.
19/13/2015CS360 AI & Robotics CS360: AI & Robotics TTh 9:25 am - 10:40 am Shereen Khoja
Introduction: Chapter 1
Course Overview  What is AI?  What are the Major Challenges?  What are the Main Techniques?  Where are we failing, and why?  Step back and look at.
Artificial Intelligence: An Introduction Definition of AI Foundations of AI History of AI Advanced Techniques.
CSC4444: Artificial Intelligence Fall 2011 Dr. Jianhua Chen Slides adapted from those on the textbook website.
Psy Introduction1 Introduction Psychology 612.
CNS 4470 Artificial Intelligence. What is AI? No really what is it? No really what is it?
Artificial Intelligence CS 363 Kawther Abas Lecture 1 Introduction 5/4/1435.
Introduction to Artificial Intelligence and Soft Computing
1 CS 2710, ISSP 2610 Foundations of Artificial Intelligence introduction.
ARTIFICIAL INTELLIGENCE 2009 Ira Pohl TIM Oct 29, 2009.
Types of Artificial Intelligence & Pioneers in the Field By Vernon Crowder.
AI: Can Machines Think? Juntae Kim Department of Computer Engineering Dongguk University.
The Dream of an Intelligent Machine Hans W. Guesgen Computer Science Department.
Introduction to Artificial Intelligence CS 438 Spring 2008 Today –AIMA, Chapter 1 –Defining AI Next Tuesday –Intelligent Agents –AIMA, Chapter 2 –HW: Problem.
FOUNDATIONS OF ARTIFICIAL INTELLIGENCE
Artificial Intelligence Lecture 2 Department of Computer Science, International Islamic University Islamabad, Pakistan.
Princess Nora University Artificial Intelligence CS 461 Level 8 1.
Artificial Intelligence Chapter 1 –Part 1 Artificial Intelligence Dr.Hassan Al-Tarawneh.
Vilalta/Norvig/Eick: History of AI Origins McCulloch and Pitts (1943) Model of Artificial Neurons. Donald Hebb (1949) Hebbian Learning Conference at Dartmouth.
Chapter 1. Introduction in Creating Brain-like intelligence, Sendhoff et al. Course: Robots Learning from Humans Bae, Eun-bit Otology Laboratory Seoul.
CSE 473: Artificial Intelligence
Artificial Intelligence
“... artificial intelligence [AI] is the science of making machines do things that would require intelligence if done by [humans]” (Minsky, 1963) ARTIFICIAL.
Open Informatics a computer science program bachelor and master guarantor: prof. Michal Pěchouček.
CSC 290 Introduction to Artificial Intelligence
Artificial Intelligence introduction(2)
Introduction to Artificial Intelligence and Soft Computing
Introduction to Artificial Intelligence
What is AI?.
AI and Agents CS 171/271 (Chapters 1 and 2)
Future of Artificial Intelligence
CS 404 Artificial Intelligence
Introduction to Artificial Intelligence Instructor: Dr. Eduardo Urbina
Introduction to Artificial Intelligence
Artificial Intelligence Andrea Danyluk For FPS 2/2/04.
Presentation transcript:

Artificial Intelligence and interdisciplinarity Bert Kappen Symposium Neuroscience Oktober 2012

22

33

Wat is intelligentie? –Evolutionair: hoe is het ontstaan? –Principieel: hoe kan het bestaan? –Praktisch: hoe maak je het?

"Een rol van intelligentie is om onze waarnemingen aan te vullen waar deze tekort schieten, en is het gevolg van de beperkingen van onze zintuigen en van de onzekerheid in de wereld om ons heen." 1. Evolutionair perspectief

Intelligentie is een vorm van patroonherkenning

Is intelligent gedrag verenigbaar met de wetten van de natuur? 2. Principieel perspectief

Plenz Chialvo 2010

Complexe dynamische systemen zijn deterministisch en onvoorspelbaar Menselijk gedrag is mechanisch en onvoorspelbaar

Intelligent gedrag vereist complexe berekeningen en efficiente algoritmes Patroonherkenning, redeneren, leren, control,….. 3. Praktisch perspectief

Aristotle Syllogic (Aristotle, 350 BC) 'All men are mortal, and no gods are mortal, therefore no men are gods.'

From Lull (1274) to Leibniz (1670) "It is unworthy of excellent men to lose hours like slaves in the labor of calculation, which could be safely relegated to anyone else if machines were used."

Talking heads (Albertus Magnus, 13 th C.)

Descartes ( ) Dualism: proposed to place the rational mind separate from the body.

AI as an engineering problem Can we obtain the result of 10 8 years of evolution by design or learning?

The digital age

Unification of AI

Chess playing as search (Shannon, 1950) Logic Theorist (Newell and Simon, 1956) Learning checkers player (Samuel, 1952) Intelligent machines

General purpose search (brute force) not only faces complexity, but also gives non-sense results: "the spirit is willing but the flesh is weak" "the wodka is good but the meat is rotten" Expert systems

The brain is a universal Turing Machine (McCulloch & Pitts, 1943) Neural networks Multi-layered networks of the ’80s could learn anything (given sufficient data) Many successful applications

p(ab)=p(a|b)p(b) p(a)+p(not a)=1 Modern AI uses probability theory Bayes rule: p(b|a)p(a)=p(a|b)p(b)

disease=yes,no test=yes,no d t Bayes’rule p(d=1)=0.01 p(d=0)=0.99 p(t=1|d=1)=0.95 p(t=0|d=1)=0.05 P(t=1|d=0)=0.05 p(t=0|d=0)=0.05 A disease has prevalence of 1 %. A test has an accuracy of 95 % John does the test and the result is positive. What is the probability that John has the disease? P(d=1|t=1) = p(t=1|d=1) p(d=1)/p(t=1) p(t=1|d=1) p(d=1)=0.95*0.01= p(t=1)=p(t=1|d=0)p(d=0)+p(t=1|d=1)p(d=1)=0.05* *0.01=0.059 p(d=1|t=1)=0.0095/0.059=0.16

Bayes Rule Learning: –X parameters –Y training data p(x|y)=p(y|x)p(x)/p(y)

Bayes Rule Inference: –X diagnoses –Y patient findings p(x|y)=p(y|x)p(x)/p(y)

Bayes Rule Localization: –X locations –Y images database PCA Off-line x y p(x|y)=p(y|x)p(x)/p(y)

Graphical models What are probabilities given evidence: Intractable for large number of variables: 2 n for binary variables

Unification of AI

Most interesting problems are hard 101 sec sec 3015 year year year Complexity

Methods from physics help out

32

Control as inference Probe the system with uncontrolled trajectories Choose the ones that are most successful Steer according to their initial direction Improve probing and iterate

34

35

36 Result after 100 trials of motor babbling No model assumed

37 Application in robotics 37

38 Summary Research on intelligence is interdisciplinary: –Computer science, physics, neuroscience, engineering, robotics, statistics, mathematics Real progress is hard: –Hard mono-disciplinary problems Complexity of computation, “what does the brain compute?” –Interdisciplinary paradigm clashes and resolutions Bayesian revolution, Control as Statistical physics 38

39 Interdisciplinary research Interdisciplinary research profits from low hanging fruit Society wants quick results Mono-disciplinary ‘Deep and slow’ Established research paradigm Inter-disciplinary ‘Shallow and fast’ no established research paradigm

40 Interdisciplinary research The role of industry –Industry has limited vision of fundamental research (top sectoren beleid) –Participation of companies in publicly funded research sometimes confuses ‘relevance for society’ with ‘relevance for the company’. 40