New User Special Price Expires in

Let's log you in.

Sign in with Facebook


Don't have a StudySoup account? Create one here!


Create a StudySoup account

Be part of our community, it's free to join!

Sign up with Facebook


Create your account
By creating an account you agree to StudySoup's terms and conditions and privacy policy

Already have a StudySoup account? Login here


by: Vito Kilback

IntroductiontoArtificialIntelligence CS510

Marketplace > Drexel University > ComputerScienence > CS510 > IntroductiontoArtificialIntelligence
Vito Kilback
GPA 3.88


Almost Ready


These notes were just uploaded, and will be ready to view shortly.

Purchase these notes here, or revisit this page.

Either way, we'll remind you when they're ready :)

Preview These Notes for FREE

Get a free preview of these Notes, just enter your email below.

Unlock Preview
Unlock Preview

Preview these materials now for free

Why put in your email? Get access to more of this material and other relevant free materials for your school

View Preview

About this Document

Class Notes
25 ?




Popular in Course

Popular in ComputerScienence

This 37 page Class Notes was uploaded by Vito Kilback on Wednesday September 23, 2015. The Class Notes belongs to CS510 at Drexel University taught by RachelGreenstadt in Fall. Since its upload, it has received 60 views. For similar materials see /class/212434/cs510-drexel-university in ComputerScienence at Drexel University.

Similar to CS510 at Drexel

Popular in ComputerScienence


Reviews for IntroductiontoArtificialIntelligence


Report this Material


What is Karma?


Karma is the currency of StudySoup.

You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!

Date Created: 09/23/15
Lecture 9 AI and the Brain Rachel Greenstadt November 252008 Reminders Bayesian learning exercise due NOW Presentations and papers due next week Final here l29 6 pm Distinguished lecture by Milind Tambe l28 l I am extra credit for attending Topics 0 Reverse engineering the brain 0 Neural networks 0 StrongAl Debate What is Al Thinking like Thinking a human rationally 1 Most of Actinglikea Acting Cs5o human rationally Brains 1011 neurons of gt 20 types 1014 synapses 1ms 10ms cycle time Signals are noisy spike trains of electrical potential Axonal arborization Axon from another cell Synapse Dendrite Nucleus Synapses Cell body or Soma Chapter 20 Section 5 3 How to translate into computer terms 0 Kurzweil Moravec Merkle 0 Number of synapses and ring rate l 0I6 0 Extrapolation from retina IO392 l0I4 0 Energy to propagate nerve impulses IO395 0 Thagard molecules and hormones etc make neurons more complex IO23 0 Quantum states argument Penrose etc All Thinks Great and Small Bunnn V a 3ftquot Mum mm 1 I Hunuu M 60m NEW I Illlan W w m m m W me tunn MIPS r m v mm mm m x mm mm M w kw 959 mm me Iooo Mllllon Ellllnn T lllnn Megabytes Evolution or Com puter Power Cost MIPS pusmnn me mum mm mm 5m I Am N Nanuuvdr J E m quot mm 1 t muuon WWW 1 gm Wequot 7 1900 1920 19130 1960 1980 2000 2020 Year Intel says it can maintain Moore s law until 2029 How the One Machine Rivals the Human Brain Mhough Hu Ono Mthm qr man I nu procuunq pawquot mm on I am haunts 101111 quotum I phoneL D JRt Md all 61M Melcth Mien uuwumumuuwum IIIW Iauwhlm an Slang CPU Snood TIMDHCMO lien Comaup AM 1200 000 000 nonII cmpu39u 27000000 DI m 120000000 DWI cameras 220000000 M93 mam 64600900 DVRI 100000000 Wobcm 3300000000 Cell phones HS 000000 PDAi For several decades the computing power found in advanced Arti cial Intelligence and Robotics systems has been stuck at insect brain power of MlPSWhie computer power per dollar fe should be rose rapidly during this periodthe money available fell just as fastThe earliest days ofAI in the mid I960s were fueled by lavish postSputnik defense funding which gave access to I 0000000 supercomputers of the time In the postVietnam war days of the I970s funding declined and only I 000000 machines were available By the early I980sAI research had to settle for I 00000 minicomputers In the late I980s the available machines were 0000 workstations By the I990s much work was done on personal computers costing only a few thousand dollars Since then Al and robot brain power has risen with improvements in computer efficiency By I993 personal computers provided I0 MIPS by I995 it was 30 MIPS and in I997 it is over I00 MIPS Suddenly machines are reading text recognizing speech and robots are driving themselves cross country Moravec I997 Arti cial Neural Networks History 0 Belief that it was necessary to model underlying brain architecture for Al 0 In contrast to encoded symbolic knowledge best represented by expert systems 0 Hebb learning is altering strength of synaptic connections Neural Networks 0 Attempt to build a computation system based on the parallel architecture of brains 0 Characteristics 0 Many simple processing elements 0 Many connecctions 0 Simple messages 0 Adaptive interaction Bene ts of NN User friendly well reasonably Nonlinear Noise tolerant Many applications 0 Credit fraudassignment 0 Control Neurons Inputs either from outside or other neurons Weighted connections that correspond to synaptic ef ciency Threshold values to weight the inputs Passed through activation function to determine output Example Unit 0 Binary inputoutput T VlIfIO R s Linear threshold unit 0 Rule 3 a v W 9 W1 0 I If U U I1I Woo WII Wb gt 0 10 11 Bias unit 0 0 if woo WI Wb lt 0 Activation functions l 807 inl39 a a is a step function or threshold function b is a sigmoid function 11 6quot Changing the bias weight WW moves the threshold location Chapter 20 Section 5 5 How to Adapt 0 Perceptron Learning Rule 0 change the weight by an amount proportional to the difference between the desired output and the actual output As an equation AWi r D Yi where D is desired output andY is actual output Stop when converges Limits of Perceptrons 0 Minsky and Papert I969 0 Fails on linearly inseparable instances 0 XOR 0 linearly separable pattern space can be separated by single hyperplane Perceptrons vs Decision Trees Perceptron learning rule converges to a consistent function for any linearly separable data set Perce tron Denision 39I ee AmmumioA 990900 Proportion correct on test set 0 10 20 30 40 50 60 70 80 90100 Training set size MAJORITY on 11 inputs Perceptron Decvswn tree 4 O 10 20 30 40 50 60 7O 80 90100 Training set size RESTAURANT data Perceptron learns majority function easily DTL is hopeless DTL learns restaurant function easily perceptron cannot represent it Multilayer Perceptrons M LP Hidden gt40 0 Back Propagation Start with a set of known examples supervised approach Assign random initial weights Run examples through and calculate the mean squared error Propagate the error by making small changes to the weights at each level Lather rinse repeat Class Debate Resolved That humans will develop Strong Al intelligence that matches or exceeds human intelligence in the next IOO years Three groups PRO CON Moderators Format 5 min opening argument starting with PRO 2 min rebuttals l5 min of moderated questions 2 min closing arguments Address questions to moderators CS 5O Lecture 5 Logic and Planning Rachel Greenstadt October 28 2008 Reminders 0 Revised proposals due next week 0 Remember to add a brief section discussing progress 0 Also an election next week if you need to be late to vote okay bring sticker Midterm Stats 0 Mean 73 0 Stdev I8 0 Solutions posted soon Overview Knowledge Bases Propositional logic Exercise Planning Representations Knowledge Bases Knowledge base set of sentences in a formal language Declarative approach to building an agent TELL it what it needs to know Then it ASKs itself what to do answers follow from knowledge base Building a model of the world part from Lecture l BDI is one paradigm for this Architecture of a Knowledge Base Domain independent Inference Engine algorlthms Kn0Wedge Base Domainspeci c conte Knowledgebased agent function KB AGEXTi n si ic m returns an m Hun AB a lawn w TELLlKBJleE PEnCEr T sngvz u mmmi muuu 1w iler39IUrl3lERiI TELLBIHE Acmmszwncm mu In 14 4 1 return m imn The agent must be able to Represent states actions etc Incorporate new percepts Update internal representation of the world Deduce hidden properties of the world Deduce appropriate actions Wumpus World PAGE description 39 5 Percepts Breeze Glitter Smell 4 V 3 1 7 Actions Left turn Right turn Forward Grab Release Shoot 2 gaze 2339263 Get gold back to start 1 39 START without entering pit or wumpus square 1 2 3 4 Environment Squares adjacent to wumpus are smelly Squares adjacent to pit are breezy Glitter if and only if gold is in the same square Shooting kills the wumpus if you are facing it Shooting uses up the only arrow Grabbing picks up the gold if in the same square Releasing dr0ps the gold in the same square AIMA Slides Stuart Russell and Peter Norvig 1998 Chapter 6 Logic 0 Logics formal languages for representing information 0 Syntax de nes the sentences in the language 0 Semantics de ne the meaning of sentences Types of Logic 0 Logics are characterized by what they commit to as primitives 0 Ontological commitment What exists 0 Facts objects times beliefs Lnnpim iniiilnmi iI ummmwm Elihh39lmiiiluil 4 39mmmi lil39lli Ivmwammi hm Fimrm vi my a i miuiluiiiv11mm Fivu luqii 1x Lu h Ulljl f i li LHimh Lm mii ix ii39ldileh mum mix iwm ui lliiili n iwidiwm1umu u nwy iaiwRuiluu mi mivl I im1dwi39h lumm mi MM U ii33w i inliii39n i Entailment Kska 0 Knowledge Base KB entails a sentence 0 if and only if 0 0 is true in all worlds where KB is true 9 0 Eg the KB containing the Giants won and the Reds won entails either the Giants won or the Reds won Inference 0 KB 0 sentence 0 can be derived from KB by procedure i 0 Soundness i is sound if whenever KB 0 it is also true that KB I0 0 Completenessi is complete if whenever KB I0 it is also true that KB 0 Propositional Logic Syntax 0 Propositional logic is the simplest logic illustrates basic ideas 0 The proposition symbols Pl P2 etc are sentence If S is a sentence n5 is a sentence If 51 and 5393 is a sentence 539 3 is a sentence If 539 and 83 is a sentence 51 V39 is a sentence If 5391 and 93 is a sentence 6391 3 S is a sentence If 51 and 53 is a sentence 5 1 gt 5393 is a sentence Propositional Logic Semantics Each model specifies truefalse for each proposition symbol Eg l B Truc Truc False Rules for evaluating truth with respect to a model in vS is true iii 539 is false 5 5 3 is true iff 5 1 is true and 5393 is true 6391 V 53 is true ifl Si is true or 52 is true SI 9 S is true ifl SI is false g 539 is true 39 e is false iff 51 is true 539 is false 1 51 ltgt S is true iff Si gt 53 is true S a 5391 is true


Buy Material

Are you sure you want to buy this material for

25 Karma

Buy Material

BOOM! Enjoy Your Free Notes!

We've added these Notes to your profile, click here to view them now.


You're already Subscribed!

Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'

Why people love StudySoup

Bentley McCaw University of Florida

"I was shooting for a perfect 4.0 GPA this semester. Having StudySoup as a study aid was critical to helping me achieve my goal...and I nailed it!"

Janice Dongeun University of Washington

"I used the money I made selling my notes & study guides to pay for spring break in Olympia, Washington...which was Sweet!"

Steve Martinelli UC Los Angeles

"There's no way I would have passed my Organic Chemistry class this semester without the notes and study guides I got from StudySoup."

Parker Thompson 500 Startups

"It's a great way for students to improve their educational experience and it seemed like a product that everybody wants, so all the people participating are winning."

Become an Elite Notetaker and start selling your notes online!

Refund Policy


All subscriptions to StudySoup are paid in full at the time of subscribing. To change your credit card information or to cancel your subscription, go to "Edit Settings". All credit card information will be available there. If you should decide to cancel your subscription, it will continue to be valid until the next payment period, as all payments for the current period were made in advance. For special circumstances, please email


StudySoup has more than 1 million course-specific study resources to help students study smarter. If you’re having trouble finding what you’re looking for, our customer support team can help you find what you need! Feel free to contact them here:

Recurring Subscriptions: If you have canceled your recurring subscription on the day of renewal and have not downloaded any documents, you may request a refund by submitting an email to

Satisfaction Guarantee: If you’re not satisfied with your subscription, you can contact us for further help. Contact must be made within 3 business days of your subscription purchase and your refund request will be subject for review.

Please Note: Refunds can never be provided more than 30 days after the initial purchase date regardless of your activity on the site.