Machine Learning 101

Transcript

1. Machine Learning 101 Ali Akbar Septiandri Universitas Al Azhar Indonesia
2. Previously...
3. Cross Industry Standard Process for Data Mining (CRISP-DM)
4. Data Science Venn Diagram
5. What is the role of machine learning algorithms?
6. “Fundamentally, machine learning involves building mathematical models to help understand data.” - Jake VanderPlas
7. Tasks in Machine Learning 1. Predicting stock price 2. Differentiating cat vs. dog pictures 3. Spam identification 4. Community detection 5. Mimicking famous painting style 6. Mastering the game of go and chess 7. etc.
8. Task Categories 1. Supervised learning a. Predicting stock price b. Differentiating cat vs. dog pictures c. Spam identification 2. Unsupervised learning a. Community detection b. Mimicking famous painting style 3. Reinforcement learning a. Mastering the game of go and chess
9. - Iris Dataset - by R.A. Fisher (1936) - 4 attributes: sepal length, sepal width, petal length, petal width - 3 labels: Iris Setosa, Iris Versicolour, Iris Virginica Let’s take an example dataset...
15. Nearest Neighbour - Finding the closest reference - What does it mean by “closest”? - Humans comprehend visualisations very well - Can computers do the same?
16. At the lowest level, computers only understand 0 or 1
17. Euclidean Distance
18. Euclidean Distance
19. Are you sure?
20. 1. Find some k closest references 2. Use majority vote 3. We need to compute pairwise distances k-Nearest Neighbours
22. Conventional statistics can not do that
23. We need high computational power
24. What if we only want to see the subgroups in the data?
25. Clustering - Finding subgroups in the data - Your neighbours in the same housing complex regardless of their class - Unsupervised learning
27. k-Means Clustering
28. k-Means Clustering 1. Uses Euclidean distance as well 2. k = number of clusters 3. Centroids to represent clusters
32. Deep Learning
34. Digit Recognition MNIST Dataset
35. Classifying objects from pictures [Krizhevsky, 2009]
38. A neural network [Nielsen, 2016]
39. Logistic Regression y = σ(w 0 + w 1 x 1 )
40. Predicting traffic jams from CCTV pictures
41. Mimicking famous paintings
43. Other Machine Learning Algorithms
44. Naive Bayes
45. Decision trees
46. Linear regression with polynomial basis functions
47. “No free lunch”
48. Thank you