Introduction to Artificial Intelligence
Introduction to Artificial Intelligence COMPSCI 188
Popular in Course
Popular in ComputerScienence
This 2 page Class Notes was uploaded by Mr. Hayley Barton on Thursday October 22, 2015. The Class Notes belongs to COMPSCI 188 at University of California - Berkeley taught by Staff in Fall. Since its upload, it has received 31 views. For similar materials see /class/226660/compsci-188-university-of-california-berkeley in ComputerScienence at University of California - Berkeley.
Reviews for Introduction to Artificial Intelligence
Report this Material
What is Karma?
Karma is the currency of StudySoup.
Date Created: 10/22/15
SciFi Al Compared to Real Al Cyberdyne Systems Ali Rahimi and T8OO Series VS ShenHui Lee s Model 101 web cam VIDEO Vision Perception Object and character recognition Scene segmentation Image classificaiton Image from Erik Sudderth Robotics Robotics Part mecheng PartAl Reality much harderthan simulations Technologies Vehicles Rescue Soccer Lots of automation In his class We ignore mechanical aspects I MethOdS for plannmg Images from stanfordracingorg CMU RoboCup Honda ASIMO sites I MethOdS for contrOI Video from PieterAbbeel with J Zico Kolter and Andrew Ng Natural Language Speech technologies Automa ic speech recognition ASR Dialog systems speaker identification meeting analysis etc Language processing technologies Machine translation Question answering Linguisticanalysis Our research covers a range of topics in natural language processing Nquot 539 3 r950quot I Logic Logical systems v1 112 v3 Theorem provers I 771 V 02 V 273 NASAfaultdiagnOSIs Question answering 711 V 712 V 113 771V772V713 Methods Deduction systems Constraint satisfaction Satisfying assignment Satisfiability solvers 01 is true huge advances here 222 1s false 713 is false Game Playing May 3997 Deep Blue vs Kasparov First match won against worldchampion Intelligent creative play 200 million board positions per second Humans understood 999 of Deep Blue39s moves Can do about the same now with a big PC cluster Open question How can humans compete with computers at all 1996 Kasparov Beats Deep Blue I could feel I could smell a new kind of intelligence across the table 1997 Deep Blue Beats Kasparov Deep Blue hasn39t proven anything Text from Bart Selman image from IBM s Deep Blue pages Decis ion Making Scheduling eg Fraud detection Spam classifiers airline routing military Route planning eg mapquest Medical diagnosis Automated help desks Web search engines etc Rational Decisions We ll use the term rational in a particular way Rational maximally achieving predefined goals Rational only concerns what decisions are made not the thought process behind them Goals are expressed in terms of the utility of outcomes 0 Being rational means maximizing your expected utility A better title for this course would be Computational Rationality What Ab out the Brain Brains human minds are very good at making rational decisions but not perfect Brains are to intelligence as wings are to flight Brains aren t as modular as software Lessons learned prediction and simulation are key to decision making Motor speech area of Emma Temporal lob i Longitudin a lissui lv39ucul a jtiar39gaia rc39ilzi fobc Premotor aiea Piccc39wal gy39us Poslcunt39al gyiLs Parietal lube Occiaita ch Designing Rational Agents 0 An agent is an entity that perceives and acts Agent f o A rational agent selects Sensors 1 actions that maximize its Percepts m utility function o Characteristics of the g percepts environment and action space dictate 3 techniques for selecting Actuators Actions rational actions t K J o This course is about 0 General Al techniques for a variety of problem types o Learning to recognize when and how a new problem can be solved with an existing technique Pacman as an Agent Agent Sensors lt Percepts EnVIronment Actuators Actions httpinsteecsberkeleyeducs188pacmanhtml Reflex Agents Consider the past and present but not future predictions to select an action Encode preferences as a function of the percepts and action Agent Sensors I Preference function J V Actuators Agent trials