1 / 36

CS 445/545 Machine Learning Winter, 2014

CS 445/545 Machine Learning Winter, 2014. See syllabus at http://web.cecs.pdx.edu/~mm/ MachineLearningWinter2014 / Lecture slides will be posted on this website before each class. . Use of laptops, phones, etc. in class:

gaye
Download Presentation

CS 445/545 Machine Learning Winter, 2014

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. CS 445/545 Machine LearningWinter, 2014 See syllabus at http://web.cecs.pdx.edu/~mm/MachineLearningWinter2014/ Lecture slides will be posted on this website before each class.

  2. Use of laptops, phones, etc. in class: • Please don’t, unless you are using it to take notes or view class slides.

  3. What is machine learning? • Textbook definitions of “machine learning”: • Detecting patterns and regularities with a good andgeneralizable approximation (“model” or “hypothesis”) • Execution of a computer program to optimize the parameters of the model using training data or past experience.

  4. Training Examples: Class 1 Training Examples: Class 2 Test example: Class = ?

  5. Training Examples: Class 1 Training Examples: Class 2 Test example: Class = ?

  6. Training Examples: Class 1 Training Examples: Class 2 Test example: Class = ?

  7. Training Examples: Class 1 Training Examples: Class 2 Test example: Class = ?

  8. Training Examples: Class 1 Training Examples: Class 2 Test example: Class = ?

  9. Training Examples: Class 1 Training Examples: Class 2 Test example: Class = ?

  10. Types of machine learning tasks • Classification • Output is one of a number of classes (e.g., ‘A’) • Regression • Output is a real value (e.g., ‘$35/share”)

  11. Types of Machine Learning Methods • Supervised • provide explicit training examples with correct answers • e.g. neural networks with back-propagation • Unsupervised • no feedback information is provided • e.g., unsupervised clustering based on similarity

  12. “Semi-supervised” • some feedback information is provided but it is not detailed • e.g., only a fraction of examples are labeled • e.g., reinforcement learning: reinforcement single is single-valued assessment of current state

  13. Relation between “artificial intelligence” and “machine learning”?

  14. Key Ingredients for Any Machine Learning Method • Features (or “attributes”)

  15. From: "TopShop" <mm@pdx.edu> To: <mm@pdx.edu> Subject: Dear mm get 77% off List-Unsubscribe: <mailto:unsubscribe-5E587F53488BBC388181@harkinsroofing.com> MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Wassup mm, how are you?! Hurry up, last day huge discount... Our shop have more than 3085 pisl online. Buy more, pay less! Today TopSellers: ** Proepcia - 0.12$ ** Levitra - 1.64$ ** Ciails - 1.78$ ** Vaigra - 0.79$ Purchase here http://TvBqeUP.ngadoctor.ru/

  16. http://www.kaggle.com/c/dogs-vs-cats

  17. Data: Instances and Features Instance: Individual example of a particular category or class 6 instances of the class “B” 56 instances of the class “A”

  18. Data: Instances and Features Features: Collection of attributes of a single instance Feature Vector: N-dimensional vector describing a single instance 0 0 0 0 0 0 0 0 0 1 1 1 ...

  19. 32 x 32 Another example of computing features

  20. f1 f2 f3 f4 f5 f6 f7 f8 Another example of computing features f9 x = (f1, f2, ..., f64) = (0, 2, 13, 16, 16, 16, 2, 0, 0, ...) Etc. ..

  21. Notation for Instances and Features Instance:x (boldface vector) Set of M instances: {x1, x2, ... , xM} Instance as feature vector, with N features: x = (x1, x2, ..., xN) Instance as a point in an N-dimensional space: x x2 x3 x1

  22. Key Ingredients for Any Machine Learning Method • Features (or “attributes”) • Underlying Representation for “hypothesis”, “model”, or “target function”:

  23. Key Ingredients for Any Machine Learning Method • Features (or “attributes”) • Underlying Representation for “hypothesis”, “model”, or “target function”: • Hypothesis space

  24. Learning method

  25. Learning method • Data:

  26. Learning method • Data: • Training data • Used to train the model

  27. Learning method • Data: • Training data • Used to train the model • Validation data • Used to select model complexity, to determine when to stop training, or to alter training method

  28. Learning method • Data: • Training data • Used to train the model • Validation data • Used to select model complexity, to determine when to stop training, or to alter training method • Test data • Used to evaluate trained model

  29. Learning method • Data: • Training data • Used to train the model • Validation data • Used to select model complexity, to determine when to stop training, or to alter training method • Test data • Used to evaluate trained model • Evaluation method

  30. Assumption of all ML methods: Inductive learning hypothesis: Any hypothesis that approximates target concept well over sufficiently large set of training examples will also approximate the concept well over other examples outside of the training set. Difference between “induction” and “deduction”?

  31. This class • Perceptrons • Evaluating Classifiers • Neural Networks • Support Vector Machines • Ensemble Learning • Bayesian Learning • Hidden Markov Models • Unsupervised Learning • Hierarchical Networks and “Deep Learning” • Feature Selection • Evolutionary Learning

  32. Class Logistics • Weekly Quizzes (closed book, etc.) • 30 minutes, on Wednesdays • Final Exam (open notes, open everything) • Grading: • Homework 50% • In-Class Quizzes 30% • Final exam 20% • My office hours • Tu, Th. 2-3 or by appt. • Mailing list • ml2014@cs.pdx.edu • Readings / Slides • On-line • Homework • Four assignments • Late HW policy • Website: • http://www.cs.pdx.edu/~mm/MachineLearningWinter2014

  33. Don’t forget to ask questions!

More Related