Intro

Prehistory of AI: Philosophy (reasoning, planning, learning, science, automation), psychology (learning, cognitive models), linguistics (grammar, formal representation of meaning), Babbage designed a universal machine, “a thinking machine” for “all subjects in the universe.”

Short AI History: In the first half of the 20th century, science fiction familiarized the world with the concept of artificially intelligent robots. By the 1950s, we had a generation of scientists, mathematicians, and philosophers with the concept of artificial intelligence in their minds. One such person was Alan Turing, a young British polymath who explored the mathematical possibility of AI. AI’s “father” was John McCarthy and Claude Shannon Dartmouth. An attempt will be made to find how to make machines use language, form abstractions and concepts, solve kinds of problems now reserved for humans, and improve themselves. We think a significant advance can be made if we work together for a summer.

Timeline: 1943: McCulloch & Pitts: Boolean circuit model of the brain, 1950: Turing's “Computing Machinery and Intelligence”, the 1950s: Early AI programs: chess, checkers (RL), theorem proving, 1956: Dartmouth meeting: “Artificial Intelligence” adopted, 1965: Robinson's complete algorithm for logical reasoning, 1969—79: Early development of knowledge-based systems, 1980—88: Expert systems industry booms, 1988—93: Expert systems industry busts: “AI Winter”, Resurgence of probability, focus on uncertainty, General increase in technical depth, Agents and learning systems… “AI Spring”?, Big data, big computing, deep learning, and AI are used in many industries.

-Agent: An entity that perceives and acts

-Rational Agent: Selects actions that maximize its expected utility

Characteristics of the sensors, actuators, and environment dictate techniques for selecting rational actions

This course is about: General AI techniques for many problem types and learning to choose and apply the method appropriate for each problem

-Actuators: When an agent perceives its environment through sensors and acts upon it

-Agent Function: Maps percept sequences to actions, it’s generated by an agent program running on a machine

Example with Pacman: Performance measure ▪ -1 per step; + 10 food; +500 win; -500 die; +200 hit a scared ghost, Environment ▪ Pacman dynamics (incl ghost behavior), Actuators ▪ Left Right Up Down or NSEW, Sensors ▪ Entire state is visible (except power pellet duration).

Agent Design Environment Type: Partially observable => agent requires memory (internal state) ▪ Stochastic => agent may have to prepare for contingencies ▪ Multi-agent => agent may need to behave randomly ▪ Static => agent has time to compute a rational decision ▪ Continuous time => continuously operating controller ▪ Unknown physics => need for exploration ▪ Unknown perf. measure => observe/interact with the human principal

Finding the best path is to use optimization, followed by gradient.

Gain time and space and remove unnecessary data.

PCA (Principal Component Analysis)

SIFT feature visualization

PCA compression 144D → 3D

ICA

Noise Filtering, Auto-Encoder

Denoised image using 15 PCA components

PCA vs Fisher Linear Discriminant

Limitation of PCA

Challenge: Facial Recognition

PCA application - Eigenfaces: Generate a set of Eigenfaces