Today ays plan an What is AI? Plus a brief history CS 4100 - PowerPoint PPT Presentation

Today ay’s plan an • What is AI? Plus a brief history CS 4100 Artificial al Intelligence • What this course is about and some logistics In Instructor or: Jan-Willem van de Meent Websit ite: https://course.ccs.neu.edu/cs4100f19 At Attribution : many of these slides are modified versions of those distributed with the UC Berkeley CS188 materials Thanks to John DeNero, Dan Klein, and Peter Abbeel.

Defining Artificial al Intelligence • Precise definitions are surprisingly elusive • Informally ally: The discipline of creating intelligent algorithms • Here we’ve just offloaded the complexity into the term intelligent In press: Any algorithm that makes predictions • In • AI often means the same thing as machine learning • Most machine learning is essentially computational statistics

The Brai ain as as Inspirat ation for AI • Humans are in some case very good at making rational decisions, but certainly not perfect Brains are very hard to reverse engineer! • “ Brains are to intelligence as wings are to flight ” • Lessons learned from the brain: memory and • simulation are key to decision making

al approaches to thinking and acting Thi This Cour urse: e: Rat ational Rat ational al decisions The imperative: maximize expected utility Here we use the term rat ational al in a specific, technical manner • Rat ational al: maximally achieving pre-defined goals • Rationality only concerns what de decisions are made (not the thought processes underpinning them) • Goal als are expressed in terms of the ut utility of ou outc tcom omes • Being rat ational al means max aximizing your expected utility We may think of this view of AI as Computational Rationality

Topi To pics cs in n Thi This Co Cour urse Agent-bas Age ased Approac aches to Mak aking Decisions • We will be concerned with designing ag agents that • operate in some env environm nment ent (very broadly interpreted) Par art I: Mak aking Decisions (Acting Rationally) • maximize some notion of expected uti utility ty Par art II: Reas asoning under Uncertai ainty (Thinking Rationally) • Typically there are pe percept ptio ions available and ac actions we can take Par art III: Ethics of AI • The agent operates in stat ates (e.g., locations) Thr Throug ugho hout: ut: Applications and emerging research • The environment might be stat atic ( constant ) or dynam amic ( changing ) • It also might be fu fully lly or only par artial ally observable Vac acuum-clean aner world Vac acuum-clean aner world The agent might be a Roomba The agent might be a Roomba We have: some perceptions via sensors •

Vac acuum-clean aner world Vac acuum-clean aner world The agent might be a Roomba The agent might be a Roomba We have: We have: some perceptions via sensors some perceptions via sensors • • • actions we can take (suck, move right, …) • actions we can take (suck, move right, …) utility we’d like to maximize • (some combination of ”don’t break” and ”clean all floors”) Rat ational al vac acuum clean aner? Vac acuum-clean aner world The agent might be a Roomba We have: some perceptions via sensors • • actions we can take (suck, move right, …) utility we’d like to maximize • (some combination of ”don’t break” and ”clean all floors”) • states where are we on the floor?

Rat ational al vac acuum clean aner Rat ational al vac acuum clean aner Here actions depend deterministically on states; i.e., this is a look up table. Here actions depend deterministically on states; i.e., this is a look up table. We will call such agents reflex ag agents . We will call such agents reflex ag agents . Ques Question: n: How would we evaluate such an agent? Defining a a Good Notion of Utility is Har ard • One meas asure of performan ance: Amount of dirt cleaned in an eight-hour shift. • Pr Problem: The agent can maximize this performance by cleaning the floor, then dumping out all the dirt, and then cleaning it again. • Better meas asure of performan ance: Amount of time that floor is clean. A A brie ief sele lectiv tive his istor tory of of AI AI • Ru Rule of Thumb (from Textbook) k): It is better to design utility according to outcomes, than according to how we think the agent should behave.

A brief an and selective history of AI Min Minsk sky 1940-1950: 1940 1950: Early y da days ys • - 1943: McCulloch & Pitts: Boolean circuit model of brain - 1950: Turing's “Computing Machinery and Intelligence” Built the first Neural Network computer in 1950 Min Minsk sky The Turi The Turing ng Test Test Built the first Neural Network computer in 1950 The same year that Turing proposed his test Image credit: http://turing100.blogspot.com/2012/05/one-month-to-biggest-turing-test.html

https://xkcd.com/329/ A brief an and selective history of AI Initial al Successes: Checkers 1940-1950: 1940 1950: Early y da days ys • - 1943: McCulloch & Pitts: Boolean circuit model of brain - 1950: Turing's “Computing Machinery and Intelligence” • 1950 1950—70: 70: Exc xcitement – Lo Look, Ma, a, no han ands! - 1950s: Early AI programs, including Samuel's checkers program, Newell & Simon's Logic Theorist, Gelernter's Geometry Engine - 1956: Dartmouth meeting: “Artificial Intelligence” adopted - 1965: Robinson's complete algorithm for logical reasoning

Bold clai aims an and extreme optimism Initial al Successes: Toy Worlds an and Sear arch “ machines will be capable, within twenty years, of doing any work a man can do” - Herbert Simon, in 1965 A brief an and selective history of AI A brief an and selective history of AI 1940-1950: 1940 1950: Early y da days ys 1940-1950: 1940 1950: Early y da days ys • • - - 1943: McCulloch & Pitts: Boolean circuit model of brain 1943: McCulloch & Pitts: Boolean circuit model of brain - - 1950: Turing's “Computing Machinery and Intelligence” 1950: Turing's “Computing Machinery and Intelligence” • 1950 1950—70: 70: Exc xcitement – Lo Look, Ma, a, no han ands! • 1950 1950—70: 70: Exc xcitement – Lo Look, Ma, a, no han ands! - - 1950s: Early AI programs, including Samuel's checkers program, 1950s: Early AI programs, including Samuel's checkers program, Newell & Simon's Logic Theorist, Gelernter's Geometry Engine Newell & Simon's Logic Theorist, Gelernter's Geometry Engine - - 1956: Dartmouth meeting: “Artificial Intelligence” adopted 1956: Dartmouth meeting: “Artificial Intelligence” adopted - - 1965: Robinson's complete algorithm for logical reasoning 1965: Robinson's complete algorithm for logical reasoning 1970 1970—90: 90: Knowledg dge-ba based d appr pproaches 1970—90: 1970 90: Knowledg dge-ba based d appr pproaches • • - - 1969—79: Early development of knowledge-based systems 1969—79: Early development of knowledge-based systems - - 1980—88: Expert systems industry booms 1980—88: Expert systems industry booms - - 1988—93: Expert systems industry busts: “AI Winter” 1988—93: Expert systems industry busts: “AI Winter” • 1990—2012: 1990 2012: Statistical appr pproaches - Resurgence of probability, focus on uncertainty - General increase in technical depth - Agents and learning systems… “AI Spring”?

A brief an and selective history of AI AI / ML L is star arting to be everywhere 1940-1950: 1940 1950: Early y da days ys • - 1943: McCulloch & Pitts: Boolean circuit model of brain - 1950: Turing's “Computing Machinery and Intelligence” 1950—70: 1950 70: Exc xcitement – Lo Look, Ma, a, no han ands! • • Advertising - 1950s: Early AI programs, including Samuel's checkers program, Newell & Simon's Logic Theorist, Gelernter's Geometry Engine • Search engines - 1956: Dartmouth meeting: “Artificial Intelligence” adopted • Route planning - 1965: Robinson's complete algorithm for logical reasoning • Spam / fraud detection 1970—90: 1970 90: Knowledg dge-ba based d appr pproaches • - 1969—79: Early development of knowledge-based systems • Automated help desks - 1980—88: Expert systems industry booms • Product recommendations - 1988—93: Expert systems industry busts: “AI Winter” • 1990 1990—2012: 2012: Statistical appr pproaches - Resurgence of probability, focus on uncertainty - General increase in technical depth - Agents and learning systems… “AI Spring”? • 2012—: E 2012 : Excitement – Lo Look, Ma, a, no han ands ag agai ain? - Big data, big compute, neural networks - Some re-unification of sub-fields - AI used in many industries What at Can an AI Do? Co Computer Vision (Perception) Qu Quiz: Which of the following can be done at present? Facial Recognition Image Segmentation Pose Recognition 3-D Understanding • Play a mean game of Jeopardy? • Drive safely along a curving mountain road? • Drive safely along Huntington Avenue? • Converse successfully with another person for an hour? Play world champion level GO • Put away the dishes and fold the laundry? • [DensePose] Source: TechCrunch [Caesar et al., ECCV 2017] • Translate spoken Chinese into spoken English in real time? • Write an intentionally funny story?

Lan Languag age an and Speech https://cvdazzle.com AlphaG aGo beat ats World Cham ampion Le Lee Sedol Play aying Atar ari with Deep Q-lear arning [Mnih et al., Nature 2015]