CS110 Lesson 27 - Intro to AI

AI is the part of computer science concerned with creating systems that can:

- Think like humans - Act like humans - Think rationally - Act rationally

Define Strong AI.

A form of AI that exhibits human cognitive abilities; refers to systems that can think and accomplish tasks on their own

Define Weak AI.

A form of AI that models intelligent behavior; refers to systems that can accomplish a wide range of problems, but operate within a predefined range of functions

Data Generation Example

A random picture that's never created before

Anomaly Detection Example

A sudden, unsual spike in a graph

Partially Observable Environment

AI can only see "part of the picture"

Collaborative Environment

AI can work with other agents (they can all succeed!)

Competitive Environment

AI competes against an agent with an opposite goal

Full Observable Environment

AI has access to all of the information available

Ranking/Recommendation Example

Buying a lot of Pillow Pets on Amazon triggers more Pillow Pet advertisements on Amazon and other websites

Clustering Problem

Determine how to categorize a group of data based on their common characteristics (ex: features)

Classification Problem Example

Distinguishing pictures that have dogs vs. no dogs

Optimization Example

Figuring out the maximum amount of items you can buy with $20; Figuring out the shortest distance to drive through all 50 U.S. state capitals

Optimization

Given a function, determine what input(s) will result in the max/min value

Ranking/Recommendation (estimate or prediction)

Given an input (ex: search query, list of previous actions) determine what action is most likely to occur next

Capabilities of Weak AI

Limited to no ability to think for themselves, can only perform tasks for which they were designed to accomplish, can (sometimes) learn when provided with training data

Capabilities of Strong AI

Logical thinking, can make decisions/plan in uncertain conditions, self-learn, communicate in a natural language

Clustering Problem Example

Organizing into groups those who are interested in the product and targeting them with more advertising, compared to ignoring those who are not interested in the product

Data Generation

Produce novel data in order to achieve a specified goal

Continuous Environment

Real-time; driving; every little instant determines the result

What does it mean if the human questioner is unable to determine if the responses came from another person or a computer?

The AI is an AI and passed the Turing Test

How does an AI pass the Turing Test?

The human questioner needs to not know which person/thing responded to it

Deterministic Environment

The outcome of an action is known with 100% certainty

Stochastic Environment

The outcome of an action is known with some probability

Discrete Environment

Turn-based; checkers; your move/turn determines the result

Classification Problem

Using a set of training examples as a guide, categorize new inputs as belonging to one or more categories

Describe the Turing Test.

a test of a machine's ability to exhibit intelligent behavior equivalent to, or indistinguishable from, that of a human

Anomaly Detection

the process of identifying rare or unexpected items or events in a data set that do not conform to other items in the data set