CSC 4510 - M.A. Papalaskari - Villanova University Lecture 2: History and Overview of Machine Learning CSC 4510 – Machine Learning Dr. Mary-Angela Papalaskari Department of Computing Sciences Villanova University Course website: www.csc.villanova.edu/~map/4510/ CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University “It won’t truly be an autonomous vehicle until you instruct it to drive to work and it heads to the beach instead.” Brad Templeton, Software designer and a consultant for the Google project on Autonomous Vehicles NYTimes 1/24/12 http://www.nytimes.com/2012/01/24/technology/googles-autonomous-vehicles-draw-skepticism-at-legal-symposium.html?_r=2&nl=technology&emc=techupdateema22 CSC 4510 - M.A. Papalaskari - Villanova University

What are the goals of AI research? Artifacts that THINK like HUMANS Artifacts that THINK RATIONALLY ACT HUMAN == TURING TEST, LOEBNER PRIZE THINK HUMAN == COG SCI THINK RATIONALLY == LOGICIST, LAWS OF THOUGHT ACT RATIONALLY == BOOK APPROACH, MORE GENERAL, MAXIMIZE PROBABILITY OF ACHIEVING GOAL Artifacts that ACT RATIONALLY Artifacts that ACT like HUMANS CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University A Bit of History Arthur Samuel (1959) wrote a program that learnt to play checkers well enough to beat him. CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Advances in mathematical logic, information theory, concept of neural computation 1943: McCulloch & Pitts Neuron 1948: Shannon: Information Theory 1949: Hebbian Learning cells that fire together, wire together 1950s Early computers. Dartmouth conference coins the phrase “artificial intelligence” and Lisp is proposed as the AI programming language 1950: Turing Test 1956: Dartmouth Conference 1958: Friedberg: Learn Assembly Code 1959: Samuel: Learning Checkers CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University A.I. funding increased (mainly military). Famous quote: “Within a generation ... the problem of creating 'artificial intelligence' will substantially be solved.” Early symbolic reasoning approaches. Logic Theorist, GPS, Perceptrons 1969: Minsky & Papert “Perceptrons” 1970s A.I. “winter” – Funding dries up as people realize this is a hard problem! Limited computing power and dead-end frameworks lead to failures. eg: Machine Translation Failure CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Rule based “expert systems” used in medical / legal professions. Bio-inspired algorithms (Neural networks, Genetic Algorithms). Again: A.I. promises the world – lots of commercial investment Expert Systems (Mycin, Dendral, EMYCIN Knowledge Representation and reasoning: Frames, Eurisko, Cyc, NMR, fuzzy logic Speech Recognition (HEARSAY, HARPY, HWIM) ML: 1982: Hopfield Nets, Decision Trees, GA & GP. 1986: Backpropagation, Explanation-Based Learning CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Some concrete successes begin to emerge. AI diverges into separate fields: Computer Vision, Automated Reasoning, Planning systems, Natural Language processing, Machine Learning… …Machine Learning begins to overlap with statistics / probability theory. 1992: Koza & Genetic Programming 1995: Vapnik: Support Vector Machines CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University First commercial-strength applications: Google, Amazon, computer games, route-finding, credit card fraud detection, spam filters, etc… Tools adopted as standard by other fields e.g. biology CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Using machine learning to detect spam emails. To: you@gmail.com GET YOUR DIPLOMA TODAY! If you are looking for a fast and cheap way to get a diploma, this is the best way out for you. Choose the desired field and degree and call us right now: For US: 1.845.709.8044 Outside US: +1.845.709.8044 "Just leave your NAME & PHONE NO. (with CountryCode)" in the voicemail. Our staff will get back to you in next few days! ALGORITHM Naïve Bayes Rule mining CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Using machine learning to recommend books. ALGORITHMS Collaborative Filtering Nearest Neighbour Clustering CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Using machine learning to identify faces and expressions. ALGORITHMS Decision Trees Adaboost CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Using machine learning to identify vocal patterns ALGORITHMS Feature Extraction Probabilistic Classifiers Support Vector Machines + many more…. CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University ML for working with social network data: detecting fraud, predicting click-thru patterns, targeted advertising, etc etc etc . ALGORITHMS Support Vector Machines Collaborative filtering Rule mining algorithms Many many more…. CSC 4510 - M.A. Papalaskari - Villanova University

Samuel’s definition of ML is still relevant Arthur Samuel (1959). Machine Learning: Field of study that gives computers the ability to learn without being explicitly programmed. CSC 4510 - M.A. Papalaskari - Villanova University

Tom Mitchell (1998): Well-posed Learning Problem A computer program is said to learn from experience E with respect to some task T and some performance measure P, if its performance on T, as measured by P, improves with experience E. CSC 4510 - M.A. Papalaskari - Villanova University

Defining the Learning Task Improve on task, T, with respect to performance metric, P, based on experience, E. T: Playing checkers P: Percentage of games won against an arbitrary opponent E: Playing practice games against itself T: Recognizing hand-written words P: Percentage of words correctly classified E: Database of human-labeled images of handwritten words T: Driving on four-lane highways using vision sensors P: Average distance traveled before a human-judged error E: A sequence of images and steering commands recorded while observing a human driver. T: Determine which students like oranges or apples P: Percentage of students’ preferences guessed correctly E: Student attribute data CSC 4510 - M.A. Papalaskari - Villanova University

Designing a Learning System Choose the training experience Choose exactly what is too be learned, i.e. the target function. Choose a learning algorithm to infer the target function from the experience. A learning algorithm will also determine a performance measure Learner Environment/ Experience Knowledge Performance Element CSC 4510 - M.A. Papalaskari - Villanova University

Improve on task, T, with respect to Quick check: Improve on task, T, with respect to performance metric, P, based on experience, E. Suppose your email program watches which emails you do or do not mark as spam, and based on that learns how to better filter spam. What is the task T in this setting? Watching you label emails as spam or not spam. Classifying emails as spam or not spam The number (or fraction) of emails correctly classified as spam/not spam. None of the above—this is not a machine learning problem. CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Machine learning Supervised Learning Classification Regression Unsupervised learning Others: Reinforcement learning, recommender systems. Also talk about: Practical advice for applying learning algorithms. CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Machine learning Supervised Learning Classification Regression Unsupervised learning Others: Reinforcement learning, recommender systems. Also talk about: Practical advice for applying learning algorithms. CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Classification Example: Credit scoring Differentiating between low-risk and high-risk customers from their income and savings Discriminant: IF income > θ1 AND savings > θ2 THEN low-risk ELSE high-risk CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Classification Example: Iris data 4 attributes sepal length sepal width petal length petal width Differentiating between 3 different types of iris CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Iris Data more plots: CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Classification Tree CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Face Recognition Training examples of a person Test images ORL dataset, AT&T Laboratories, Cambridge UK CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Housing price prediction. Price ($) in 1000’s Size in feet2 Supervised Learning “right answers” given Regression: Predict continuous valued output (price) CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Machine learning Supervised Learning Classification Regression Unsupervised learning Others: Reinforcement learning, recommender systems. Also talk about: Practical advice for applying learning algorithms. CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Regression Example: Price of a used car x : car attributes y : price y = g (x | q ) g ( ) model, q parameters y = wx+w0 CSC 4510 - M.A. Papalaskari - Villanova University

Regression Applications Navigating a car: Angle of the steering Kinematics of a robot arm CSC 4510 - M.A. Papalaskari - Villanova University

Supervised Learning: Uses Prediction of future cases: Use the rule to predict the output for future inputs Knowledge extraction: The rule is easy to understand Compression: The rule is simpler than the data it explains Outlier detection: Exceptions that are not covered by the rule, e.g., fraud CSC 4510 - M.A. Papalaskari - Villanova University

Quick check: You’re running a company, and you want to develop learning algorithms to address each of two problems. Problem 1: You have a large inventory of identical items. You want to predict how many of these items will sell over the next 3 months. Problem 2: You’d like software to examine individual customer accounts, and for each account decide if it has been hacked/compromised. Should you treat these as classification or as regression problems? Treat both as classification problems. Treat problem 1 as a classification problem, problem 2 as a regression problem. Treat problem 1 as a regression problem, problem 2 as a classification problem. Treat both as regression problems.

CSC 4510 - M.A. Papalaskari - Villanova University Machine learning Supervised Learning Classification Regression Unsupervised learning Others: Reinforcement learning, recommender systems. Also talk about: Practical advice for applying learning algorithms. CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Supervised Learning x2 x1 CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Unsupervised Learning x2 x1 CSC 4510 - M.A. Papalaskari - Villanova University

Unsupervised Learning Learning “what normally happens” No output Clustering: Grouping similar instances Example applications Customer segmentation Image compression: Color quantization Bioinformatics: Learning motifs CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Genes Individuals CSC 4510 - M.A. Papalaskari - Villanova University [Source: Su-In Lee, Dana Pe’er, Aimee Dudley, George Church, Daphne Koller]

Organize computing clusters Social network analysis Image credit: NASA/JPL-Caltech/E. Churchwell (Univ. of Wisconsin, Madison) Astronomical data analysis image obtained from NASA website. http://www.nasa.gov/multimedia/imagegallery/image_feature_874.html RCW 79 is seen in the southern Milky Way, 17,200 light-years from Earth in the constellation Centaurus. The bubble is 70-light years in diameter, and probably took about one million years to form from the radiation and winds of hot young stars. The balloon of gas and dust is an example of stimulated star formation. Such stars are born when the hot bubble expands into the interstellar gas and dust around it. RCW 79 has spawned at least two groups of new stars along the edge of the large bubble. Some are visible inside the small bubble in the lower left corner. Another group of baby stars appears near the opening at the top. NASA's Spitzer Space Telescope easily detects infrared light from the dust particles in RCW 79. The young stars within RCW79 radiate ultraviolet light that excites molecules of dust within the bubble. This causes the dust grains to emit infrared light that is detected by Spitzer and seen here as the extended red features. Image credit: NASA/JPL-Caltech/E. Churchwell (Univ. of Wisconsin, Madison) -------------------- From NASA use guidelines: http://www.nasa.gov/multimedia/guidelines/index.html Using NASA Imagery and Linking to NASA Web Sites 10.13.05 Still Images, Audio Files and Video NASA still images, audio files and video generally are not copyrighted. You may use NASA imagery, video and audio material for educational or informational purposes, including photo collections, textbooks, public exhibits and Internet Web pages. This general permission extends to personal Web pages. This general permission does not extend to use of the NASA insignia logo (the blue "meatball" insignia), the retired NASA logotype (the red "worm" logo) and the NASA seal. These images may not be used by persons who are not NASA employees or on products (including Web pages) that are not NASA sponsored. If the NASA material is to be used for commercial purposes, especially including advertisements, it must not explicitly or implicitly convey NASA's endorsement of commercial goods or services. If a NASA image includes an identifiable person, using the image for commercial purposes may infringe that person's right of privacy or publicity, and permission should be obtained from the person. Any questions regarding application of any NASA image or emblem should be directed to: Photo Department NASA Headquarters 300 E St. SW Washington, DC 20546 Tel: (202)358-1900 Fax: (202)358-4333 Linking to NASA Web Sites NASA Web sites are not copyrighted, and may be linked to from other Web sites, including individuals' personal Web sites, without explicit permission from NASA. However, such links may not explicitly or implicitly convey NASA's endorsement of commercial goods or services. NASA images may be used as graphic "hot links" to NASA Web sites, provided they are used within the guidelines above. This permission does not extend to use of the NASA insignia, the retired NASA logotype or the NASA seal. Restrictions Please be advised that: 1) NASA does not endorse or sponsor any commercial product, service, or activity. 2) The use of the NASA name, initials, any NASA emblems (including the NASA insignia, the NASA logo and the NASA seal) which would express or imply such endorsement or sponsorship is strictly prohibited. 3) Use of the NASA name or initials as an identifying symbol by organizations other than NASA (such as on foods, packaging, containers, signs, or any promotional material) is prohibited. 4) NASA does permit the use of the NASA logo and insignia on novelty and souvenir-type items. However, such items may be sold and manufactured only after a proposal has been submitted to and approved by a Visual Identity representative from the Public Outreach Division (Phone: 202/358-1750) in accordance with 14 CFR (Code of Federal Regulations) Part 1221. Permission is granted on a nonexclusive basis as it is not NASA's policy to grant exclusive rights to use any of the agency identities. 5) No approval for use is authorized by NASA when the use can be construed as an endorsement by NASA of a product, service or activity. 6) NASA emblems should be reproduced only from original reproduction proofs, transparencies, or computer files available from NASA Headquarters. Please be advised that approval must be granted by a Visual Identity representative from the Public Outreach Division ( Tel: 202/358-1750) before any reproduction materials can be obtained. Market segmentation Astronomical data analysis CSC 4510 - M.A. Papalaskari - Villanova University

Quick check: Of the following examples, which would you address using an unsupervised learning algorithm? (Check all that apply.) Given email labeled as spam/not spam, learn a spam filter. Given a database of customer data, automatically discover market segments and group customers into different market segments. Given a set of web pages found on the web, automatically detect the ones that are syllabi for AI or software engineering courses Given a dataset of patients diagnosed as either having diabetes or not, learn to classify new patients as having diabetes or not. Given a database of nutrition data, automatically discover categories of food items.

CSC 4510 - M.A. Papalaskari - Villanova University Machine learning Supervised Learning Classification Regression Unsupervised learning Others: Reinforcement learning, recommender systems. Also talk about: Practical advice for applying learning algorithms. CSC 4510 - M.A. Papalaskari - Villanova University

Reinforcement Learning Learning a policy: A sequence of outputs No supervised output but delayed reward Credit assignment problem Game playing Robot in a maze Multiple agents, partial observability, ... CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Machine learning Supervised Learning Classification Regression Unsupervised learning Others: Reinforcement learning, recommender systems. Also talk about: Practical advice for applying learning algorithms. CSC 4510 - M.A. Papalaskari - Villanova University

Supervised or Unsupervised learning? Iris Data

CSC 4510 - M.A. Papalaskari - Villanova University Summary ML grew out of work in AI Optimize a performance criterion using example data or past experience. Types of learning Supervised Unsupervised Role of Statistics: Inference from a sample Role of Computer science: Data representation and modeling Efficient algorithms to solve optimization problems Representing and evaluating the model for inference CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Resources: Datasets UCI Repository: http://www.ics.uci.edu/~mlearn/MLRepository.html UCI KDD Archive: http://kdd.ics.uci.edu/summary.data.application.html Statlib: http://lib.stat.cmu.edu/ Delve: http://www.cs.utoronto.ca/~delve/ CSC 4510 - M.A. Papalaskari - Villanova University

CSC 4510 - M.A. Papalaskari - Villanova University Resources: Journals Journal of Machine Learning Research www.jmlr.org Machine Learning Neural Computation Neural Networks IEEE Transactions on Neural Networks IEEE Transactions on Pattern Analysis and Machine Intelligence Annals of Statistics Journal of the American Statistical Association ... CSC 4510 - M.A. Papalaskari - Villanova University

Resources: Conferences International Conference on Machine Learning (ICML) European Conference on Machine Learning (ECML) Neural Information Processing Systems (NIPS) Uncertainty in Artificial Intelligence (UAI) Computational Learning Theory (COLT) International Conference on Artificial Neural Networks (ICANN) International Conference on AI & Statistics (AISTATS) International Conference on Pattern Recognition (ICPR) ... Some of the slides in this presentation are adapted from: Prof. Frank Klassner’s ML class at Villanova the University of Manchester ML course http://www.cs.manchester.ac.uk/ugt/COMP24111/ The Stanford online ML course http://www.ml-class.org/ CSC 4510 - M.A. Papalaskari - Villanova University