Artificial Intelligence

Definition of Artificial Intelligence

  • Webster’s Dictionary - Intelligence: The ability to learn and solve problems.
  • Wikipedia - Artificial Intelligence is the intelligence exhibited by machines or software.
  • McCarthy - The science and engineering of making intelligent machines.
  • Russel and Novig (AI book) - The study and design of intelligent agents, where an intelligent agent is a system that perceives its environment and takes actions that maximize its chances of success.

Why Artificial Intelligence?

Andrew Ng - Just as the Industrial Revolution freed up a lot of humanity from physical drudgery, I think AI has the potential to free humanity from a lot of mental drudgery.

What is AI?

Four schools of thoughts (Russel and Norvig)

  1. Thinking Humanly - cognitive approach

  2. Thinking Rationally - Law of thoughts

  3. Acting Humanly - mimicking human’s action

Turing test (Alan Turing 1950)

Major Components of AI: Knowledge, Reasoning, Language, Understanding, Learning.

  1. Acting Rationally - expected to maximize goal achievement, given the available information.

Rational Agent - one that acts so as to achieve the best outcome, or when there is uncertainty, the best expected outcome.

Applications of AI

 1. Speech Recognition
 2. Handwriting recognition
 3. Machine translation
 4. Robotics
 5. Recommendations system (Collaborative Filtering)
 6. Search Engines
 7. Email
 8. Face detection
 9. Chess

  1. Autonomous Driving

Foundation of AI

  1. Mathematics
  2. Philosophy
  3. Linguistics
  4. Psychology
  5. Sociology
  6. Economics
  7. Computer Science

Search Agents

  • agents that work towards a goal
  • consider the impact of actions on future states
  • agent’s job is to identify the action or series of actions that lead to the goal
  • Two kinds of search: Uninformed and Informed Search.

Uninformed Search

use no domain knowledge

Uninformed Search Methods

  1. Depth- First Search
  2. Breadth- First Search
Reflex Agent
  • Consider how the world is
  • Do not consider the future consequences of their actions
  • Choose action based on current perceptive(& maybe memory)
Planning Agent
  • Ask “what if”
  • Decision based on (hypothesized) consequences of actions
  • Must have a model of how the world evolves in response to their actions
  • Must formulate a goal (test)
  • Consider how the world would be

Search Problems

  • consists of

  
  1. A state space
  2. a successor or function (with actions, costs)
  3. A start state and a goal state

A solution is a sequence of actions ( a plan) which transforms the start state to a goal state

State Space Graphs
  • A mathematical representation of a search problem
  • Each state only occurs one
Search Trees
  • tree of plans and their outcomes
  • the start state is the root node
  • Children correspond to successors
  • Nodes show states, but corresponds to plan that achieve those states
Depth-First Search
  • Vertically before horizontally
  • Keyword: Pop and Stack

Breadth-First Search

  • Horizontally before vertically
  • keyword: queue