Download to read offline
![1_Introduction to Machine Learning [Autosaved].pptx](https://cdn.slidesharecdn.com/ss_thumbnails/1introductiontomachinelearningautosaved-250910004933-3913b711-thumbnail.jpg?width=640&height=640&fit=bounds)
An Introduction to Machine Learning.pptx
- 1. Machine Learning ML YusufUzun Aksaray University P. hD. Student Seminar Presentation
- 2. 2 Machine Learning Introduction •Humankind used machines to suport work force. • Nowadays, humankind uses machines to support brain power owing to the development of Machine Learning!
- 3.
- 4. Machine Learning Definiton MachineLearning explores the study and construction of algorithims that • can learn from past data • can make future predictions based on learnt data • can improve their performance through experience using training data • Machine Learning is a set of Computational Programs used to develop statistical models • Typically, an algorithm has a number of parameters whose values are learnt from the data • The ability to perform a task in a situation which has never been encountered before (Learning = Generalization)
- 5. Machine Learning Evolution •ML is actually getting computers program themselves • Programming ise the bottleneck, then let the data be used to construct a model that would do the work
- 6. Machine Learning Importance •“Machine learning is going to result in a real revolution” (Greg Papadopoulos, CTO, Sun) • “Machine learning is the next Internet” (Tony Tether, Former Director, DARPA) • “A breakthrough in machine learning would be worth ten Microsofts” (Bill Gates, Microsoft) • “Web rankings today are mostly a matter of machine learning” (Prabhakar Raghavan, Former Dir. Research, Yahoo) • “Machine learning is the hot new thing” (John Hennessy, President, Stanford) • “Machine learning today is one of the hottest aspects of computer science” (Steve Ballmer, CEO, Microsoft) • “Machine learning is today’s discontinuity” (Jerry Yang, Founder, Yahoo)
- 7. Machine Learning Mission •Needed and developped to solve complicated problems that are very hard or impossible to program by hand • We don’t know what program to write because we don’t know how our brain does it. • Even if we had a good idea about how to do it, the program might be horrendously complicated. • Instead of writing a program by hand, we collect lots of examples that specify the correct output for a given input • A machine learning algorithm then takes these examples and produces a program that does the job • Discover new useful knowledge from large databases (data mining) • Ability to mimic human and replace certain monotonous tasks that require some intelligence • Develop systems that can automatically adapt and customize themselves to individual users
- 8. Why now? • Floodof available big data (especially with the advent of the Internet) • Increasing computational power • Growing progress in available algorithms and theory developed by researchers • Increasing support from industries 8
- 9. Sample Applications • Patternrecognition • Speech recognition, Natural language processing • Web search engines • Recommendation systems • Medical Diagnosis • Computational biology • Finance & Credit • Fraud detection • E-commerce • Space exploration • Robotics, unassisted control of a vehicle • Information extraction • Social networks • …………………
- 10. The machine learningframework • Apply a prediction function to a feature representation of the image to get the desired output: f( ) = “apple” f( ) = “tomato” f( ) = “cow”
- 11. The machine learningframework y = f(x) • Training: given a training set of labeled examples {(x1,y1), …, (xN,yN)}, estimate the prediction function f by minimizing the prediction error on the training set • Testing: apply f to a never before seen test example x and output the predicted value y = f(x) output prediction function Image feature
- 12.
- 13. Using a validationset Divide the total dataset into three subsets: • Training data is used for learning the parameters of the model. • Test data is used to get a final, unbiased estimate of how well the network works. We expect this estimate to be worse than on the validation data.
- 14. Algorithm Stages Every machinelearning algorithm has three components: – Representation – Evaluation – Optimization
- 15. Generalization • Underfitting: modelis too “simple” to represent all the relevant class characteristics • Overfitting: model is too “complex” and fits irrelevant characteristics (noise) in the data
- 16. Types of Learning •Supervised (inductive) learning – Training data includes desired outputs • Unsupervised learning – Training data does not include desired outputs • Reinforcement learning - The learner interacts with the world via “actions” and tries to find an optimal policy of behavior with respect to “rewards” it receives from the environment
- 17. • Learning topredict a discrete value from a predefined set of values (classification) • Learning to predict a continuous/real value (regression) • Group similar data points together (clustering) • Find lower-dimensional manifold preserving some properties of the data (dimension reduction) Types of Learning
- 18. Linear Classification Find alinear function to separate the classes:
- 19. Classification • Decision rulesdivide input space into decision regions separated by decision boundaries
- 20. 20 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 Model
- 21. 21 Classification: Applications • Creditrisk assessment • Pattern recognition • Email Spam Filtering • Face recognition • Optical Character Recognition: Different handwriting styles • Speech recognition • Medical disease diagnosis: From symptoms to illnesses
- 23. 23 Prediction: Regression • Example:Price of a used car • x : car attributes y : price y = g (x ) g ( ) model y = wx+w0
- 24. 24 Regression: Applications • PredictingGold & Stock Future Prices • Predict the Price of a Used Car • Automatic Steering
- 25. 25 Unsupervised Learning • Thechallenging part here is to identify whether the data can be separated into groups that are relatively distinct • Training data contains inputs but does not include outputs • Learning “what happens in normal life” • Clustering: Grouping similar instances • Example applications • Customer segmentation in CRM • Individualised marketing • Image compression
- 26.
- 28.
- 29. Collaborative Filtering predict howwell a user will like an item that he has not rated given a set of historical preference judgments for a community of users
- 30.
- 31. 31 Reinforcement Learning • Nosupervised output but delayed reward • Applications: • Game playing • Robot in a maze • Multiple agents
- 32. Decision Trees Decisiontree A tree with nodes representing condition testing and leaves representing classes Decision list If condition 1 then class 1 elseif condition 2 then class 2 elseif …. Sample Algorithms: ID3, C4.5
- 33.