1. Machine Learning Overview Tamara Berg CS 590-133 Artificial Intelligence Many slides throughout the course adapted from Svetlana Lazebnik, Dan Klein, Stuart Russell, Andrew Moore, Percy Liang, Luke Zettlemoyer, Rob Pless, Killian Weinberger, Deva Ramanan 1

2. Announcements HW4 is due April 3 Reminder: Midterm2 next Thursday –Next Tuesday’s lecture topics will not be included (but material will be on the final so attend!) Midterm review –Monday, 5pm in FB009

3. Midterm Topic List Be able to define the following terms and answer basic questions about them: Reinforcement learning –Passive vs Active RL –Model-based vs model-free approaches –Direct utility estimation –TD Learning and TD Q-learning –Exploration vs exploitation –Policy Search –Application to Backgammon/Aibos/helicopters (at a high level) Probability –Random variables –Axioms of probability –Joint, marginal, conditional probability distributions –Independence and conditional independence –Product rule, chain rule, Bayes rule

4. Midterm Topic List Bayesian Networks General –Structure and parameters –Calculating joint and conditional probabilities –Independence in Bayes Nets (Bayes Ball) Bayesian Inference –Exact Inference (Inference by Enumeration, Variable Elimination) –Approximate Inference (Forward Sampling, Rejection Sampling, Likelihood Weighting) –Networks for which efficient inference is possible Naïve Bayes –Parameter learning including Laplace smoothing –Likelihood, prior, posterior –Maximum likelihood (ML), maximum a posteriori (MAP) inference –Application to spam/ham classification –Application to image classification (at a high level)

5. Midterm Topic List HMMs –Markov Property –Markov Chains –Hidden Markov Model (initial distribution, transitions, emissions) –Filtering (forward algorithm) Machine Learning –Unsupervised/supervised/semi-supervised learning –K Means clustering –Training, tuning, testing, generalization

6. Machine learning Image source: https://www.coursera.org/course/mlhttps://www.coursera.org/course/ml

7. Machine learning Definition –Getting a computer to do well on a task without explicitly programming it –Improving performance on a task based on experience

8. Big Data!

9. What is machine learning? Computer programs that can learn from data Two key components –Representation: how should we represent the data? –Generalization: the system should generalize from its past experience (observed data items) to perform well on unseen data items.

10. Types of ML algorithms Unsupervised –Algorithms operate on unlabeled examples Supervised –Algorithms operate on labeled examples Semi/Partially-supervised –Algorithms combine both labeled and unlabeled examples

12. Clustering –The assignment of objects into groups (aka clusters) so that objects in the same cluster are more similar to each other than objects in different clusters. –Clustering is a common technique for statistical data analysis, used in many fields, including machine learning, data mining, pattern recognition, image analysis and bioinformatics.

13. Euclidean distance, angle between data vectors, etc

15. K-means clustering Want to minimize sum of squared Euclidean distances between points x i and their nearest cluster centers m k

29. Source: Hinrich Schutze

30. Hierarchical clustering strategies Agglomerative clustering Start with each data point in a separate cluster At each iteration, merge two of the “closest” clusters Divisive clustering Start with all data points grouped into a single cluster At each iteration, split the “largest” cluster

31. P Produces a hierarchy of clusterings P P P

32. P

33. Divisive Clustering Top-down (instead of bottom-up as in Agglomerative Clustering) Start with all data points in one big cluster Then recursively split clusters Eventually each data point forms a cluster on its own.

34. Flat or hierarchical clustering? For high efficiency, use flat clustering (e.g. k means) For deterministic results: hierarchical clustering When a hierarchical structure is desired: hierarchical algorithm Hierarchical clustering can also be applied if K cannot be predetermined (can start without knowing K) Source: Hinrich Schutze

35. Clustering in Action – example from computer vision

36. Recall: Bag of Words Representation  Represent document as a “bag of words”

37. Bag-of-features models Slides adapted from Fei-Fei Li, Rob Fergus, and Antonio Torralba

38. Bags of features for image classification 1.Extract features

39. 2.Learn “visual vocabulary” Bags of features for image classification

40. 1.Extract features 2.Learn “visual vocabulary” 3.Represent images by frequencies of “visual words” Bags of features for image classification

41. … 1. Feature extraction

42. 2. Learning the visual vocabulary …

43. Clustering …

44. 2. Learning the visual vocabulary Clustering … Visual vocabulary

45. Example visual vocabulary Fei-Fei et al. 2005

46. 3. Image representation ….. frequency Visual words

47. Types of ML algorithms Unsupervised –Algorithms operate on unlabeled examples Supervised –Algorithms operate on labeled examples Semi/Partially-supervised –Algorithms combine both labeled and unlabeled examples

52. Example: Sentiment analysis http://gigaom.com/2013/10/03/stanford-researchers-to-open-source-model-they-say-has-nailed-sentiment-analysis/ http://nlp.stanford.edu:8080/sentiment/rntnDemo.html

53. Example: Image classification apple pear tomato cow dog horse inputdesired output

54. http://yann.lecun.com/exdb/mnist/index.html

55. Example: Seismic data Body wave magnitude Surface wave magnitude Nuclear explosions Earthquakes

57. The basic classification framework y = f(x) Learning: given a training set of labeled examples {(x 1,y 1 ), …, (x N,y N )}, estimate the parameters of the prediction function f Inference: apply f to a never before seen test example x and output the predicted value y = f(x) outputclassification function input

58. Naïve Bayes classifier A single dimension or attribute of x

59. Example: Image classification Car Input: Image Representation Classifier (e.g. Naïve Bayes, Neural Net, etc Output: Predicted label

60. Example: Training and testing Key challenge: generalization to unseen examples Training set (labels known)Test set (labels unknown)

62. Some classification methods 10 6 examples Nearest neighbor Shakhnarovich, Viola, Darrell 2003 Berg, Berg, Malik 2005 … Neural networks LeCun, Bottou, Bengio, Haffner 1998 Rowley, Baluja, Kanade 1998 … Support Vector Machines and Kernels Conditional Random Fields McCallum, Freitag, Pereira 2000 Kumar, Hebert 2003 … Guyon, Vapnik Heisele, Serre, Poggio, 2001 …

63. Classification … more soon

64. Types of ML algorithms Unsupervised –Algorithms operate on unlabeled examples Supervised –Algorithms operate on labeled examples Semi/Partially-supervised –Algorithms combine both labeled and unlabeled examples

65. Supervised learning has many successes recognize speech, steer a car, classify documents classify proteins recognizing faces, objects in images... Slide Credit: Avrim Blum

66. However, for many problems, labeled data can be rare or expensive. Unlabeled data is much cheaper. Need to pay someone to do it, requires special testing,… Slide Credit: Avrim Blum

67. However, for many problems, labeled data can be rare or expensive. Unlabeled data is much cheaper. Speech Images Medical outcomes Customer modeling Protein sequences Web pages Need to pay someone to do it, requires special testing,… Slide Credit: Avrim Blum

68. However, for many problems, labeled data can be rare or expensive. Unlabeled data is much cheaper. [From Jerry Zhu] Need to pay someone to do it, requires special testing,… Slide Credit: Avrim Blum

69. Need to pay someone to do it, requires special testing,… However, for many problems, labeled data can be rare or expensive. Unlabeled data is much cheaper. Can we make use of cheap unlabeled data? Slide Credit: Avrim Blum

70. Semi-Supervised Learning Can we use unlabeled data to augment a small labeled sample to improve learning? But unlabeled data is missing the most important info!! But maybe still has useful regularities that we can use. But… Slide Credit: Avrim Blum