CS 445/545 Machine Learning Winter, 2012
description
Transcript of CS 445/545 Machine Learning Winter, 2012
![Page 1: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/1.jpg)
CS 445/545 Machine Learning
Winter, 2012
• Course overview:
– Instructor• Melanie Mitchell
– Textbook• Machine Learning: An Algorithmic Approach by
Stephen Marsland
![Page 2: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/2.jpg)
– Homework• Bi-weekly assignments, programming or using existing
code.
• Do on your own; no teamwork, but you can discuss general ideas with classmates
– Quizzes• Short quiz (20 minutes) every Wednesday (starting
1/25) to test basic understanding of the material presented in class and in the readings.
• If you must miss a quiz, see me to schedule a make-up quiz.
– Final exam• Take-home open-book exam.
![Page 3: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/3.jpg)
– Grading• Homework: 50% • In-class quizzes: 30%• Final: 20%
– Mailing list: [email protected](please give your e-mail address on the sign-up sheet)
![Page 4: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/4.jpg)
• Academic Integrity– You can discuss concepts with other students, but all
work you hand in must be your own.
– You may not copy any text from other students, papers, the internet, or any other source, without clearly referencing it.
– See syllabus for URL with PSU policy.
![Page 5: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/5.jpg)
• Late homework policy:– If needed, request an extension before the assignment is
due. Otherwise, 5% of the assignment grade will be subtracted for each day the homework is late.
![Page 6: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/6.jpg)
• Use of laptops, phones, etc. in class:
– Please don’t, unless you are using it to take notes or view class slides.
![Page 7: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/7.jpg)
What is machine learning?
• Textbook definitions of “machine learning”:
– Detecting patterns and regularities with a good and generalizable approximation (“model” or “hypothesis”)
– Execution of a computer program to optimize the parameters of the model using training data or past experience.
![Page 8: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/8.jpg)
Training Examples: Class 1 Training Examples: Class 2
Test example: Class = ?
![Page 9: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/9.jpg)
Training Examples: Class 1 Training Examples: Class 2
Test example: Class = ?
![Page 10: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/10.jpg)
Training Examples: Class 1 Training Examples: Class 2
Test example: Class = ?
![Page 11: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/11.jpg)
Training Examples: Class 1 Training Examples: Class 2
Test example: Class = ?
![Page 12: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/12.jpg)
Training Examples: Class 1 Training Examples: Class 2
Test example: Class = ?
![Page 13: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/13.jpg)
Training Examples: Class 1 Training Examples: Class 2
Test example: Class = ?
![Page 14: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/14.jpg)
Any observations from these machine learning examples?
![Page 15: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/15.jpg)
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”)
![Page 16: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/16.jpg)
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
![Page 17: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/17.jpg)
• “Semi-supervised”
– feedback information is provided, but is not detailed
• examples:
– e.g., reinforcement learning: reinforcement single is single-valued assessment of current state
![Page 18: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/18.jpg)
Relation between “artificial intelligence” and “machine learning”?
![Page 19: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/19.jpg)
Key Ingredients for Any Machine Learning Method
• Features (or “attributes”)
• Underlying representation for “hypothesis”, “model”, or “target function”:
• Hypothesis space:
![Page 20: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/20.jpg)
• Learning method:
• Data: Divide into three parts. – 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
![Page 21: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/21.jpg)
Demonstration
• What is:– Representation:
– Features:
– Hypothesis space:
– Learning method:
– Data:
– Evaluation method:
• Notion of “confusion matrix”
![Page 22: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/22.jpg)
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”?
![Page 23: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/23.jpg)
What questions does ML ask?
• Which is the best method for a given learning problem?
• What can be proved about convergence, performance?
• How much training data or feedback is needed for good generalization performance?
![Page 24: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/24.jpg)
• How to select/design training examples or feedback?
• How can prior knowledge be used to guide learning?
• How can meta-learning be used? (I.e., learner automatically modifies its representation, selection of training examples, etc.)
![Page 25: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/25.jpg)
Syllabus
![Page 26: CS 445/545 Machine Learning Winter, 2012](https://reader036.fdocuments.net/reader036/viewer/2022081507/56816386550346895dd471c0/html5/thumbnails/26.jpg)
Homework 1