Presentation is loading. Please wait.

Presentation is loading. Please wait.

Artificial Intelligence

Similar presentations


Presentation on theme: "Artificial Intelligence"— Presentation transcript:

1 Artificial Intelligence
CSCI 101

2 Artificial Intelligence
The study of computer systems that attempt to model and apply the intelligence of the human mind. (CSI)

3 The Turing Test A test to decide if a computer is truly intelligent.
Can a computer fool a human into convincing it that it is a human. In 1991 they began running the Turing Test annually, $100,000 will be given to the first team who can design a computer who can pass the test. It has not been one yet.

4 Eliza An early AI Psychotherapist
Works with Pattern Matching and some basic rules

5 Search Trees Search trees are a way to organize moves in a game, such that the computer player knows where to move next.

6 Xs and Os No Yes Go in a corner If Opposite Corner Free
Go in a free corner Go to opposite corner If there are 2 Xs and a space in a line If there are 2 Xs and a space in a line

7 Expert Systems

8 Expert Systems In everyday life Experts are people who have a strong knowledge base in their field, and who we go to for advice: Doctor Lawyer Piolet Weather Forecaster

9 Expert System Hold a Large Knowledge Base on a particular subject
The System is accessed when we are looking for a answers on problems.

10 Gardening Expert System Rules (CSI)
if(current date – last treatment date < 30) - no treatment if(season = winter) - no bugs if(lawn has large bare spots) - apply turf-building treatment if(sparse and not weeds) - then apply basic fertilizer if(bugs and no bare spots) - apply bug killing treatment ….

11 Using the Gardening System (CSI)
System: Does the lawn have large, bare areas? User: No System: Does the lawn show evidence of bugs? System: Is the lawn generally thin? User: Yes System: Does the lawn contain significant weeds? System: You should apply a weed-killing and fertilizer combination treatment

12 Neural Networks An Artificial Neural Network: in a computer attempts to mimic the actions of the neural networks of the human body. As time goes on the network is trained to better understand it’s situation and respond appropriately.

13 Natural Language Processing
The goal is to be able to converse with a computer as if it were a human. E.g. Siri or Alexa Three basic components: Voice recognition: Recognizing human words Natural Language Comprehension: Interpreting human communication Voice Synthesis: Recreating human speech (csi)

14 Voice Synthesis Recreating human speech. Two approaches: Dynamic Voice
Recorded Speech

15 Dynamic Voice Generation
Generation: based on putting different sounds together to make words and phrases This can end up sounding choppy, and mechanical Human language varies a lot

16 Recorded Speech Prerecorded words or phases.
Issue here is it can be hard to create a large enough vocabulary. Also, words and the ways the sound can very greatly depending on the context

17 Voice Recognition The computer needs to understand what the human is saying in order to hold a conversation This is difficult because people sound very different. Sometimes we talk fast and words are merged together. e.g. “ice cream” can sound like “I scream” It can be helpful when a human trains a computer to understand how they sound.

18 Natural Language Comprehension
The computer not only has to understand the words a person is saying, but also what they mean. Words and phrases can be ambiguous. e.g. The following can be misinterpreted: “I saw the Grand Canyon flying to New York” “The brick fell on the computer but it is not broken”

19 Robotics Two different types: Fixed robots: Assembly Line
Mobile Robots: Move and interact with their surroundings.

20 Mobile Robots Here the robot follows a set of rules when determining where to move, e.g. Wander randomly If obstacle in way If left is clear – go left else If right is clear – go right else if back is clear – go back else stop


Download ppt "Artificial Intelligence"

Similar presentations


Ads by Google