Powered by

# 1. Introduction

# What is Intelligence?

  • A wish-list of general characteristics of intelligence
    • Perception: manipulation, interpretation of data provided by sensors
    • Action: control, and use of effectors to accomplish a variety of tasks
    • Reasoning: adapting behavior to better cope with changing environments, discovery of patterns, learning to reason, plan, and act
    • Communication: with other intelligent agents including humans using signals, signs, icons, · · ·
    • Planning: formulation of plans – sequences or agenda of actions to accomplish externally or internally determined goals . . .

# What is Artificial Intelligence (AI)?

  • The exciting new effort to make computers think.. machines with minds
  • AI is the art of creating machines that perform functions that require intelligence when performed by humans
  • AI is the study of the computations that make it possible to perceive, reason, and act
  • AI is the enterprise of design and analysis of intelligent agents
    • Weak AI: machines could act as if they ere intelligent
    • Strong AI: machines that do so are actually consciously thinking (not just simulating thinking); shifted to “human-level” or “general” AI that can solve an arbitrarily wide variety of tasks, and do so as well as a human

image
image

  • Are you concerned with thought process/reasoning or behavior?
  • Do you want to model humans or measure against an ideal concept of intelligence, rationality?

# Acting humanly: Turing Test

  • Alan Turing (1950) “Computing machinery and intelligence”:
  • “Can machines think?” → “ Can machines behave intelligently?”
  • Operational test for intelligent behavior: the Imitation Game
    • img
  • Predicted that by 2000, a machine might have a 30% chance of fooling a lay person for 5 minutes
  • Annual Loebner prize competition (since 1990): the first prize of $100,000 to be awarded to the first program that passes the "unrestricted" Turing test
  • Suggested major components of AI: knowledge, reasoning, language understanding, learning, etc.

# Thinking humanly: Cognitive Science

  • 1960s “cognitive revolution”: information-processing psychology replaced prevailing orthodoxy of behaviorism
  • Cognitive science brings together computer models from AI and experimental techniques from psychology to construct precise and testable theories of human mind
  • AI and cognitive science fertilize each other
    • Incorporation of neurophysiological evidence into computational models in computer vision
    • Combination of neuroimaging methods with machine learning techniques led to the beginnings of a capacity to “read minds” (i.e. to ascertain the semantic content of a person’s inner thoughts), shed further light on how human cognition works

# Thinking rationally: Laws of Thought

  • Aristotle: what are correct arguments/thought processes?
  • Several Greek schools developed various forms of logic: notation and rules of derivation for thoughts
  • Direct line through philosophy, mathematics and logic to AI, so-called logicist
  • Problems:
    • Not all intelligent behavior is mediated by logical deliberation
    • What is the purpose of thinking? What thoughts should I have out of all the thoughts (logical or otherwise) that I could have?

# Acting rationally: Rational agent

  • This course is about designing rational agents
  • An agent is an entity that perceives and acts
  • Rational behavior: doing the right thing
  • The right thing: that which is expected to maximize goal achievement, given the available information
  • A rational agent is one that acts so as to achieve the best outcome
  • Caveat: computational limitations make perfect rationality unachievable → design best program for given machine resources

# Foundations

  • Philosophy: logic, methods of reasoning, mind as physical system, foundations of learning, language, rationality
  • Mathematics: formal representation and proof, algorithms, computation, (un)decidability, (in)tractability, probability
  • Economics: formal theory of rational decisions
  • Neuroscience: physical substrate for mental activity
  • Psychology: experimental techniques (psychophysics, etc.), behaviorism (percept/stimulus and action/response), cognitive psychology/science (views brain as info-processing device)
  • Computer engineering: building efficient computers
  • Control theory and cybernetics: homeostatic systems, stability, simple optimal agent designs
  • Linguistics: knowledge representation, grammar

# Brief history of AI

  • 1943:
    • McCulloch & Pitts: model of artificial neurons
  • 1950:
    • Turing’s “Computing Machinery and Intelligence” introduced Turing test, ML, GA, RL
  • 1956:
    • McCarthy, Minsky, Rochester, Shannon, et al., Dartmouth workshop: “Artificial Intelligence” adopted
  • 1952-69:
    • Early enthusiasm, great expectations, optimism fueled by early success on some problems thought to be hard (e.g. theorem proving);
    • NN flourished – Hebb’s learning and its enhancement in Widrow’s adalines, Rosenblatt’s perceptrons with convergence theorem
  • 1965:
    • Robinson’s complete algorithm for logical reasoning (resolution)
  • 1966-73:
    • Collapse in AI research: Progress was slower than expected;
    • Unrealistic predictions, Herbert Simon (1957) chess champion in 10 years;
    • AI discovers computational complexity;
    • NN research almost disappears
  • 1969-79:
    • Early development of knowledge-based systems
  • 1980-88:
    • Expert systems industry booms
  • 1988-93:
    • Expert systems industry busts: “AI Winter”
  • 1985-95:
    • Neural networks return to popularity
  • 1988–:
    • Resurgence of probability; general increase in technical depth, “Nouvelle AI”: ALife, GAs, soft computing
  • Mid 1990s–present: The emergence of intelligent agents in various applications
    • information retrieval
    • data mining and knowledge discovery
    • customized software systems
    • smart devices (e.g. homes, automobiles)
    • agile manufacturing systems
    • autonomous vehicles
    • bioinformatics
    • internet tools: search engines, recommender systems
    • . . .
  • 2001–: Big data
  • 2011–: Deep learning

# State of the art

AI100 st Stanford produces AI Index (aiindex.org) tracking progress

  • Robotic vehicles:
    • DARPA Grand Challenge and Urban Challenge(2005, 2007);
    • Waymo passed landmark of 10 million miles on public roads (2018), followed by commercial robotic taxi service
  • Legged locomotion:
    • Boston Dynamics’ BigDog (2008) resembles an amimal;
    • Atlas walks, jumps, and backflips (2016)
  • Autonomous planning and scheduling:
    • EUROPA planning toolkit for daily operation of NASA’s Mars rovers;
    • SAXTANT system for autonomous navigation in deep space;
    • DARPA’s DART (Dynamic Analysis and Replanning Tool) for automated logistic planning and scheduling for transportation, deployed during the Gulf war (1991);
    • Dynamic driving directions provided by Uber and Google Maps
  • Machine translation:
    • online MT systems produce adequate results
  • Speech recognition:
    • human-level, Alexa, Siri, Cortana, Google
  • Recommendation:
    • Amazon, Facebook, Netflix, Spotify, YouTube, Walmart, Coupang, . . .
  • Game playing:
    • Deep Blue defeated the world chess champion Garry Kasparov (1997);
    • Chinook defeated human checkers champion (1994), can’t lose at checkers (2007);
    • The IBM supercomputer Watson beat human champions on ‘Jeopardy!’ (2011);
    • AlphaGo/AlphaGoZero/AlphaZero (for Go, chess, shogi)/AlphaStar beat human champions
  • Image understanding:
    • ImageNet object recognition task, image captioning, etc.
  • Medicine: disease diagnosis with multimodal data
  • Climate science, . . .

© Copyright 2020 - 2022 KyuHwan Shim, All rights reserved.