DOC PREVIEW
Princeton COS 116 - Self-improvement for dummies

This preview shows page 1-2-24-25 out of 25 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 25 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 25 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 25 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 25 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 25 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

Self-improvement for dummies(Machine Learning)COS 1164/27/2006Instructor: Sanjeev AroraArtificial Intelligence! Definition of AI (Merriam-Webster):1. The capability of a machine to imitate intelligent human behavior2. Branch of computer science dealing with the simulation of intelligent behavior in computers! Definition of Learning:" To gain knowledge or understanding of or skill in by study, instruction, or experienceToday:(Next time)Lecture organization! Brief look at learning in humans/animals! Brief look at human brain! Brief look at how today’s machines learnCaveat: imitating nature may not be best strategy! Examples:BirdsAirplanesvsCheetahsRace carsvsIntelligence in animal worldIs an ant intelligent?! Build huge, well-structured colonies organized using chemical-based messaging (“Super-organism”)What about dogs?Deep mystery: How do higher animals (including humans) learn?How doesturn intoA crude first explanation: Behaviorism [Pavlov 1890’s, Skinner 1930’s]! Animals and humans can be understood in a “black box”way as a sum total of all direct conditioning events! Bell # “Food is coming” #Salivate! “This person likes me more if I call her “Mama”and that one likes me more if I call him “Papa”.Aside: What does behaviorism imply for societal organization?More thoughts on behaviorismOriginal motivation: “Can’t look inside the working brain anyway, so theory thatassumes anything about its workingis not scientific or testable.”Today,…Gives little insight into how to design machines with }intelligence. How did dogs, rats, humans sort through sensory experiences to understand reward/punishment?Chomsky’s influential critique of Behaviorism [1957]! “Internal mental structures crucial for learning.”Evidence: universal linguistic rules (“Chomsky grammars”); “self-correction” in language learning, ability to appreciate puns.1. Brain is “prewired” for language.2. Must understand mental structures to understand behaviorPresenting:Your brainThe brain! Network of 100 billion neurons! Evidence of timing mechanisms (“clock”)! About 100 firings per second" Total of 1013firings (“operations”) per second " Number of operations per sec in fast desktop PC: 1010A comparisonYour brain1011neuronsYour life on a DVD4.3 Gb for 3 hours> 1017bytes for entire lifeConclusion: Brain must contain structures that compress information and store it in an interconnected way for quick associations and retrievalA simplistic model of neurons—Neural Net [McCulloch – Pitts 1943]! Neuron computes “thresholds”! Take the sum of weights of all neighbors that are firing! If sum > T, fireInputsOutputs T: “threshold value”wi: “weight”assigned to i inputw1w2wkDoes a neural network model remind you of something??Why AI is feasible in principle: the simulation argument! Write a simulation program that simulates all 1011neurons in the brain and their firings.! For good measure, also simulates underlying chemistry, blood flow, etc.! Practical difficulty: How to figure out properties (threshold value, wi’s) of each of 1010neurons, the intricate chemistryRest of the lecture: Some Principles of machine learningOnly hope: brain is organized around simpler principles.A machine’s “experience” of the world! n sensors, each produces a number“experience” = an array of n numbers! Example: video camera: 480 x 640 pixelsn = 480 × 640 = 307200! In practice, reduce n via compression or preprocessingExample: Representing wood samplesBrownness scale 1 … 10Texture scale 1 … 10(3, 7) = wood that is fairly light brown but kind of on the rough sidelight darksmooth roughA learning task and its mathematical formulation! Given: 100 samples of oak, maple! Figure out labeling(“clustering”)! Given a new sample, classify it as oak, maple, or mahoganycolortextureoakmaple“Clustering”New pointAn algorithm to produce 2 clusters! Some notions:" Mean of k points (x1, y1), (x2, y2), ... , (xk, yk)is (“center of gravity”)" Distance between points (x1, y1), (x2, y2) is(x1– x2)2+ (y1– y2)2++++++kyyykxxxkk ...,... 21212-means Algorithm (cont.)Start by randomly breaking points into 2 clustersRepeat many times:{" Compute means of the current two clusters, say(a, b), (c, d)" Reassign each point to the cluster whose mean is closest to it; this changes the clustering}What about learning a more dynamic object?! Speech?! Motion?! Handwriting?Similar datarepresentationOne major idea: modeling uncertainty using probabilities! Example: Did I just hear“Ice cream” or “I scream”?! Assign probability ½ to each! Listen for subsequent phoneme" If “is”, use knowledge of usage patterns to increase probability of “Ice cream” to 0.9Probabilities + states = Markovianmodels! Markov decision process! Hidden Markov modelsAre “learnt” by machine after extensive training.(Condensed representation of data corpus)Rough overview of speech recognition! Markovian model of language (machine’s idea ofhow language is produced)! Estimate model parameters using data corpus + user trainingFinite state machine with probabilities on the transitionsNext lecture: Turing test! Turinghub.com! Randomly assigns you to chat with a machine or a human! Note: Machine cannot possibly store answers to all possible 5-minute


View Full Document

Princeton COS 116 - Self-improvement for dummies

Documents in this Course
Lecture 5

Lecture 5

15 pages

lecture 7

lecture 7

22 pages

Lecture

Lecture

32 pages

Lecture

Lecture

16 pages

Midterm

Midterm

2 pages

Lecture

Lecture

23 pages

Lecture

Lecture

21 pages

Lecture

Lecture

24 pages

Lecture

Lecture

22 pages

Lecture

Lecture

28 pages

Lecture

Lecture

21 pages

Lecture

Lecture

50 pages

Lecture

Lecture

19 pages

Lecture

Lecture

28 pages

Lecture

Lecture

32 pages

Lecture

Lecture

23 pages

Lecture

Lecture

21 pages

Lecture

Lecture

19 pages

Lecture

Lecture

22 pages

Lecture

Lecture

21 pages

Logic

Logic

20 pages

Lab 7

Lab 7

9 pages

Lecture

Lecture

25 pages

Lecture 2

Lecture 2

25 pages

lecture 8

lecture 8

19 pages

Midterm

Midterm

5 pages

Lecture

Lecture

26 pages

Lecture

Lecture

29 pages

Lecture

Lecture

40 pages

Lecture 3

Lecture 3

37 pages

lecture 3

lecture 3

23 pages

lecture 3

lecture 3

20 pages

Lecture

Lecture

21 pages

Lecture

Lecture

24 pages

Lecture

Lecture

19 pages

Load more
Download Self-improvement for dummies
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Self-improvement for dummies and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Self-improvement for dummies 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?