cs189 lecture 1

Preview:

Citation preview

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 1/113

CS189/CS289AIntroduction to Machine Learning

Lecture 1: Overview

Alexei Efros and Peter Bartlett

January 20, 2015

1/37

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 2/113

Organizational Issues

2/37

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 3/113

Organizational Issues

Instructors: Peter Bartlett and Alyosha Efros.

2/37

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 4/113

Organizational Issues

Instructors: Peter Bartlett and Alyosha Efros.

GSIs: Peter Gao, Yun Park, Faraz Tavakoli, Kevin Tee, Pat Virtue,Christopher Xie, Daniel Xu, Yuchen Zhang.

2/37

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 5/113

Organizational Issues

Instructors: Peter Bartlett and Alyosha Efros.

GSIs: Peter Gao, Yun Park, Faraz Tavakoli, Kevin Tee, Pat Virtue,Christopher Xie, Daniel Xu, Yuchen Zhang.

Discussion sections: You choose. If the room is full, please go toanother one. (If necessary, we may offer some specialtysections—watch website for announcements.)

2/37

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 6/113

Organizational Issues

Instructors: Peter Bartlett and Alyosha Efros.

GSIs: Peter Gao, Yun Park, Faraz Tavakoli, Kevin Tee, Pat Virtue,Christopher Xie, Daniel Xu, Yuchen Zhang.

Discussion sections: You choose. If the room is full, please go toanother one. (If necessary, we may offer some specialtysections—watch website for announcements.)

Office hours: see web site.

2/37

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 7/113

Organizational Issues

Instructors: Peter Bartlett and Alyosha Efros.

GSIs: Peter Gao, Yun Park, Faraz Tavakoli, Kevin Tee, Pat Virtue,Christopher Xie, Daniel Xu, Yuchen Zhang.

Discussion sections: You choose. If the room is full, please go toanother one. (If necessary, we may offer some specialtysections—watch website for announcements.)

Office hours: see web site.

http://www-inst.eecs.berkeley.edu/∼cs189

bCourses (+ piazza, kaggle), office hours, syllabus, assignments,readings, lecture slides, announcements.

2/37

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 8/113

Organizational Issues

Assessment:CS189   Homework 40%

Implementation and application of methods. (Kaggle)

Mathematical/reinforcement of concepts.Seven total.Late policy: 5 slip days total. That’s it.

Midterm 20%(Thursday, March 19, in the lecture slot.)Final Exam 40%

3/37

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 9/113

Organizational Issues

Assessment:

CS289A   Plus a project:

Homework 40%Midterm 20%Final Exam 20%Final Project 20%(due Friday, May 1. Proposal due Friday, April 3.)

4/37

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 10/113

Organizational Issues

(Real) Prerequisites:Math53 (vector calculus); Math54 (linear algebra); CS70 (discretemath, probability); CS188 (more probability, decision theory).

5/37

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 11/113

Organizational Issues

(Real) Prerequisites:Math53 (vector calculus); Math54 (linear algebra); CS70 (discretemath, probability); CS188 (more probability, decision theory).

No screens in lectures. (To see why, google “laptops in class.”)

5/37

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 12/113

Organizational Issues

(Real) Prerequisites:Math53 (vector calculus); Math54 (linear algebra); CS70 (discretemath, probability); CS188 (more probability, decision theory).

No screens in lectures. (To see why, google “laptops in class.”)

Ethics:

5/37

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 13/113

Organizational Issues

(Real) Prerequisites:Math53 (vector calculus); Math54 (linear algebra); CS70 (discretemath, probability); CS188 (more probability, decision theory).

No screens in lectures. (To see why, google “laptops in class.”)

Ethics:Discussion of homework problems with other students is encouraged.

5/37

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 14/113

Organizational Issues

(Real) Prerequisites:Math53 (vector calculus); Math54 (linear algebra); CS70 (discretemath, probability); CS188 (more probability, decision theory).

No screens in lectures. (To see why, google “laptops in class.”)

Ethics:Discussion of homework problems with other students is encouraged.All homeworks must be written individually (including programmingcomponents).

5/37

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 15/113

Organizational Issues

(Real) Prerequisites:Math53 (vector calculus); Math54 (linear algebra); CS70 (discretemath, probability); CS188 (more probability, decision theory).

No screens in lectures. (To see why, google “laptops in class.”)

Ethics:Discussion of homework problems with other students is encouraged.All homeworks must be written individually (including programmingcomponents).Please read the department policy on academic dishonesty. We will be

actively checking for plagiarism.

5/37

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 16/113

Organizational Issues

(Real) Prerequisites:Math53 (vector calculus); Math54 (linear algebra); CS70 (discretemath, probability); CS188 (more probability, decision theory).

No screens in lectures. (To see why, google “laptops in class.”)

Ethics:Discussion of homework problems with other students is encouraged.All homeworks must be written individually (including programmingcomponents).Please read the department policy on academic dishonesty. We will be

actively checking for plagiarism.Questions: Use piazza. Public and private.

5/37

T

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 17/113

Texts

6/37

CS189 I d i M hi L i

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 18/113

CS189: Introduction to Machine Learning

7/37

CS189 I d i M hi L i

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 19/113

CS189: Introduction to Machine Learning

Machine Learning

Systems that learn to solveinformation processing problems.

7/37

CS189 I t d ti t M hi L i

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 20/113

CS189: Introduction to Machine Learning

Machine Learning

Systems that learn to solveinformation processing problems.

LearnUse experience to improve performance:data, queries, interaction, experiments

Statistical issues are central.

7/37

CS189 I t d ti t M hi L i

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 21/113

CS189: Introduction to Machine Learning

Machine Learning

Systems that learn to solveinformation processing problems.

LearnUse experience to improve performance:data, queries, interaction, experiments

Statistical issues are central.

Systems

Computational issues are also central.

Algorithms, optimization.

7/37

An Overview of Machine Learning

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 22/113

An Overview of Machine Learning

1

2

3

8/37

An Overview of Machine Learning

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 23/113

An Overview of Machine Learning

1 Problems

2

3

8/37

An Overview of Machine Learning

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 24/113

An Overview of Machine Learning

1 Problems

2 Methods3

8/37

An Overview of Machine Learning

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 25/113

An Overview of Machine Learning

1 Problems

2 Methods3 Concepts

8/37

An Overview of Machine Learning

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 26/113

An Overview of Machine Learning

1 Problems

2 Methods3 Concepts

8/37

Classification Problems (Homework)

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 27/113

Classification Problems (Homework)

Email

ESL9/37

Classification Problems (Homework)

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 28/113

Classification Problems (Homework)

ESL

10/37

Classification

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 29/113

Classification

11/37

Classification

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 30/113

Classification

microsoft.com

12/37

Classification

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 31/113

Classification

apple.com

ESL13/37

Classification

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 32/113

ISLR

14/37

Classification

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 33/113

ISLR

15/37

Classification

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 34/113

ESL16/37

Regression

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 35/113

g

ESL17/37

Regression

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 36/113

ESL18/37

Regression

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 37/113

ESL

19/37

Regression

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 38/113

ESL

20/37

Regression

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 39/113

ESL

21/37

Density Estimation

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 40/113

ESL

22/37

Density Estimation

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 41/113

ESL

23/37

Dimensionality Reduction

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 42/113

ESL

24/37

Dimensionality Reduction

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 43/113

ESL

25/37

Dimensionality Reduction

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 44/113

ESL

26/37

Clustering

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 45/113

ESL27/37

Clustering

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 46/113

28/37

Clustering

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 47/113

ESL

29/37

Clustering

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 48/113

ESL

30/37

Machine Learning Problems

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 49/113

Classification

31/37

Machine Learning Problems

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 50/113

Classification

Regression

31/37

Machine Learning Problems

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 51/113

Classification

Regression

Density estimation

31/37

Machine Learning Problems

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 52/113

Classification

Regression

Density estimation

Dimensionality reduction

31/37

Machine Learning Problems

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 53/113

Classification

Regression

Density estimation

Dimensionality reduction

Clustering

31/37

Machine Learning Problems

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 54/113

Classification

Regression

Density estimation

Dimensionality reduction

Clustering

Ranking

31/37

Machine Learning Problems

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 55/113

Classification

Regression

Density estimation

Dimensionality reduction

Clustering

Ranking

Collaborative filtering

31/37

Machine Learning Problems

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 56/113

Classification

Regression

Density estimation

Dimensionality reduction

Clustering

Ranking

Collaborative filtering

Sequential decisionproblems:

31/37

Machine Learning Problems

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 57/113

Classification

Regression

Density estimation

Dimensionality reduction

Clustering

Ranking

Collaborative filtering

Sequential decisionproblems:

bandits

31/37

Machine Learning Problems

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 58/113

Classification

Regression

Density estimation

Dimensionality reduction

Clustering

Ranking

Collaborative filtering

Sequential decisionproblems:

banditscontextual bandits

31/37

Machine Learning Problems

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 59/113

Classification

RegressionDensity estimation

Dimensionality reduction

Clustering

Ranking

Collaborative filtering

Sequential decisionproblems:

banditscontextual banditsdynamic pricing

31/37

Machine Learning Problems

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 60/113

Classification

RegressionDensity estimation

Dimensionality reduction

Clustering

Ranking

Collaborative filtering

Sequential decisionproblems:

banditscontextual banditsdynamic pricingreinforcement learning

31/37

An Overview of Machine Learning

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 61/113

1 Problems

2 Methods

3 Concepts

32/37

Methods

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 62/113

Linear classifiers: Perceptron

Support vector machines

Gaussian class conditionals

Logistic regression

Naive Bayes

Linear discriminant analysisLinear regression

Decision trees, regression trees

Ensemble methods

Neural networksNearest neighbor

Principal components analysis

k-means clustering

33/37

Methods

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 63/113

Linear classifiers: Perceptron

Support vector machines

Gaussian class conditionals

Logistic regression

Naive Bayes

Linear discriminant analysisLinear regression

Decision trees, regression trees

Ensemble methods

Neural networksNearest neighbor

Principal components analysis

k-means clustering

1 Classification2 Regression

33/37

Methods

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 64/113

Linear classifiers: Perceptron

Support vector machines

Gaussian class conditionals

Logistic regression

Naive Bayes

Linear discriminant analysisLinear regression

Decision trees, regression trees

Ensemble methods

Neural networksNearest neighbor

Principal components analysis

k-means clustering

1 Probabilistic

modeling.

2 Prediction; not basedon a model.

33/37

An Overview of Machine Learning

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 65/113

1 Problems

2 Methods

3 Concepts

34/37

Concepts

P di i b bili i d li

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 66/113

1 Prediction versus probabilistic modeling.

35/37

Concepts

P di i b bili i d li

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 67/113

1 Prediction versus probabilistic modeling.2 Probabilistic modeling:

35/37

Concepts

1 P di ti b bili ti d li

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 68/113

1 Prediction versus probabilistic modeling.2 Probabilistic modeling:

Generative versus discriminative models.

35/37

Concepts

1 Prediction ers s probabilistic modeling

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 69/113

1 Prediction versus probabilistic modeling.2 Probabilistic modeling:

Generative versus discriminative models.Maximum likelihood estimation.

35/37

Concepts

1 Prediction versus probabilistic modeling

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 70/113

1 Prediction versus probabilistic modeling.2 Probabilistic modeling:

Generative versus discriminative models.Maximum likelihood estimation.Bayesian inference.

35/37

Concepts

1 Prediction versus probabilistic modeling

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 71/113

1 Prediction versus probabilistic modeling.2 Probabilistic modeling:

Generative versus discriminative models.Maximum likelihood estimation.Bayesian inference.

3 Optimization.

35/37

Concepts

1 Prediction versus probabilistic modeling

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 72/113

1 Prediction versus probabilistic modeling.2 Probabilistic modeling:

Generative versus discriminative models.Maximum likelihood estimation.Bayesian inference.

3 Optimization.

Convexity.

35/37

Concepts

1 Prediction versus probabilistic modeling

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 73/113

1 Prediction versus probabilistic modeling.2 Probabilistic modeling:

Generative versus discriminative models.Maximum likelihood estimation.Bayesian inference.

3 Optimization.

Convexity.

(Stochastic) gradient methods.

35/37

Concepts

1 Prediction versus probabilistic modeling

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 74/113

Prediction versus probabilistic modeling.2 Probabilistic modeling:

Generative versus discriminative models.Maximum likelihood estimation.Bayesian inference.

3 Optimization.

Convexity.

(Stochastic) gradient methods.Newton’s method.

35/37

Concepts

1 Prediction versus probabilistic modeling.

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 75/113

Prediction versus probabilistic modeling.2 Probabilistic modeling:

Generative versus discriminative models.Maximum likelihood estimation.Bayesian inference.

3 Optimization.

Convexity.

(Stochastic) gradient methods.Newton’s method.

4 Controlling complexity:

35/37

Concepts

1 Prediction versus probabilistic modeling.

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 76/113

Prediction versus probabilistic modeling.2 Probabilistic modeling:

Generative versus discriminative models.Maximum likelihood estimation.Bayesian inference.

3 Optimization.

Convexity.

(Stochastic) gradient methods.Newton’s method.

4 Controlling complexity:

Bias-variance/approximation-estimation trade-off.

35/37

Concepts

1 Prediction versus probabilistic modeling.

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 77/113

p g2 Probabilistic modeling:

Generative versus discriminative models.Maximum likelihood estimation.Bayesian inference.

3 Optimization.

Convexity.

(Stochastic) gradient methods.Newton’s method.

4 Controlling complexity:

Bias-variance/approximation-estimation trade-off.Regularization

35/37

Concepts

1 Prediction versus probabilistic modeling.

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 78/113

p g2 Probabilistic modeling:

Generative versus discriminative models.Maximum likelihood estimation.Bayesian inference.

3 Optimization.

Convexity.

(Stochastic) gradient methods.Newton’s method.

4 Controlling complexity:

Bias-variance/approximation-estimation trade-off.Regularization

Priors

35/37

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 79/113

Concepts

1 Prediction versus probabilistic modeling.

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 80/113

2 Probabilistic modeling:

Generative versus discriminative models.Maximum likelihood estimation.Bayesian inference.

3 Optimization.

Convexity.

(Stochastic) gradient methods.Newton’s method.

4 Controlling complexity:

Bias-variance/approximation-estimation trade-off.Regularization

Priors5 Practical issues:

Train/validate/test. Over-fitting.

35/37

Concepts

1 Prediction versus probabilistic modeling.

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 81/113

2 Probabilistic modeling:

Generative versus discriminative models.Maximum likelihood estimation.Bayesian inference.

3 Optimization.

Convexity.

(Stochastic) gradient methods.Newton’s method.

4 Controlling complexity:

Bias-variance/approximation-estimation trade-off.Regularization

Priors5 Practical issues:

Train/validate/test. Over-fitting.Resampling methods.

35/37

Overview (Part I: Bartlett)

Linear classification

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 82/113

Linear classification

36/37

Overview (Part I: Bartlett)

Linear classification

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 83/113

Linear classification

Statistical learning background

36/37

Overview (Part I: Bartlett)

Linear classification

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 84/113

Linear classification

Statistical learning background

Decision theory

36/37

Overview (Part I: Bartlett)

Linear classification

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 85/113

Linear classification

Statistical learning background

Decision theoryGenerative and discriminative models

36/37

Overview (Part I: Bartlett)

Linear classification

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 86/113

Linear classification

Statistical learning background

Decision theoryGenerative and discriminative modelsControlling complexity.

36/37

Overview (Part I: Bartlett)

Linear classification

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 87/113

Linear classification

Statistical learning background

Decision theoryGenerative and discriminative modelsControlling complexity.Resampling, cross-validation.

36/37

Overview (Part I: Bartlett)

Linear classification

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 88/113

Statistical learning background

Decision theoryGenerative and discriminative modelsControlling complexity.Resampling, cross-validation.The multivariate normal distribution.

36/37

Overview (Part I: Bartlett)

Linear classification

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 89/113

Statistical learning background

Decision theoryGenerative and discriminative modelsControlling complexity.Resampling, cross-validation.The multivariate normal distribution.

Linear regression

36/37

Overview (Part I: Bartlett)

Linear classification

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 90/113

Statistical learning background

Decision theoryGenerative and discriminative modelsControlling complexity.Resampling, cross-validation.The multivariate normal distribution.

Linear regression

Optimization

36/37

Overview (Part I: Bartlett)

Linear classification

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 91/113

Statistical learning background

Decision theoryGenerative and discriminative modelsControlling complexity.Resampling, cross-validation.The multivariate normal distribution.

Linear regression

Optimization

Linear Classification revisited

36/37

Overview (Part I: Bartlett)

Linear classification

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 92/113

Statistical learning background

Decision theoryGenerative and discriminative modelsControlling complexity.Resampling, cross-validation.The multivariate normal distribution.

Linear regression

Optimization

Linear Classification revisited

Logistic regression

36/37

Overview (Part I: Bartlett)

Linear classification

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 93/113

Statistical learning background

Decision theoryGenerative and discriminative modelsControlling complexity.Resampling, cross-validation.The multivariate normal distribution.

Linear regression

Optimization

Linear Classification revisited

Logistic regression

Linear Discriminant Analysis

36/37

Overview (Part I: Bartlett)

Linear classification

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 94/113

Statistical learning background

Decision theoryGenerative and discriminative modelsControlling complexity.Resampling, cross-validation.The multivariate normal distribution.

Linear regression

Optimization

Linear Classification revisited

Logistic regression

Linear Discriminant AnalysisSupport vector machines

36/37

Overview (Part I: Bartlett)

Linear classification

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 95/113

Statistical learning background

Decision theoryGenerative and discriminative modelsControlling complexity.Resampling, cross-validation.The multivariate normal distribution.

Linear regression

Optimization

Linear Classification revisited

Logistic regression

Linear Discriminant AnalysisSupport vector machines

Statistical learning theory

36/37

Overview (Part II: Efros)1 Memory-based/Instance-based learning

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 96/113

37/37

Overview (Part II: Efros)1 Memory-based/Instance-based learning

k-nearest-neighbor

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 97/113

37/37

Overview (Part II: Efros)1 Memory-based/Instance-based learning

k-nearest-neighborProperties of high dimensional spaces

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 98/113

Properties of high-dimensional spaces

37/37

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 99/113

Overview (Part II: Efros)1 Memory-based/Instance-based learning

k-nearest-neighborProperties of high-dimensional spaces

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 100/113

Properties of high-dimensional spaces

distance learningEfficient indexing and retrieval methods

37/37

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 101/113

Overview (Part II: Efros)1 Memory-based/Instance-based learning

k-nearest-neighborProperties of high-dimensional spaces

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 102/113

Properties of high dimensional spaces

distance learningEfficient indexing and retrieval methods

2 Decision treesClassification and regression trees

37/37

Overview (Part II: Efros)1 Memory-based/Instance-based learning

k-nearest-neighborProperties of high-dimensional spaces

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 103/113

Properties of high dimensional spaces

distance learningEfficient indexing and retrieval methods

2 Decision treesClassification and regression treesRandom Forests

37/37

Overview (Part II: Efros)1 Memory-based/Instance-based learning

k-nearest-neighborProperties of high-dimensional spaces

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 104/113

p g p

distance learningEfficient indexing and retrieval methods

2 Decision treesClassification and regression treesRandom Forests

3 Boosting

37/37

Overview (Part II: Efros)1 Memory-based/Instance-based learning

k-nearest-neighborProperties of high-dimensional spaces

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 105/113

p g p

distance learningEfficient indexing and retrieval methods

2 Decision treesClassification and regression treesRandom Forests

3 Boosting4 Neural networks / Deep Learning

37/37

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 106/113

Overview (Part II: Efros)1 Memory-based/Instance-based learning

k-nearest-neighborProperties of high-dimensional spaces

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 107/113

distance learningEfficient indexing and retrieval methods

2 Decision treesClassification and regression treesRandom Forests

3 Boosting4 Neural networks / Deep Learning

Multilayer perceptronsVariations such as convolutional nets

37/37

Overview (Part II: Efros)1 Memory-based/Instance-based learning

k-nearest-neighborProperties of high-dimensional spaces

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 108/113

distance learningEfficient indexing and retrieval methods

2 Decision treesClassification and regression treesRandom Forests

3 Boosting4 Neural networks / Deep Learning

Multilayer perceptronsVariations such as convolutional netsExamples and applications

37/37

Overview (Part II: Efros)1 Memory-based/Instance-based learning

k-nearest-neighborProperties of high-dimensional spaces

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 109/113

distance learningEfficient indexing and retrieval methods

2 Decision treesClassification and regression treesRandom Forests

3 Boosting4 Neural networks / Deep Learning

Multilayer perceptronsVariations such as convolutional netsExamples and applications

5

Unsupervised methods

37/37

Overview (Part II: Efros)1 Memory-based/Instance-based learning

k-nearest-neighborProperties of high-dimensional spaces

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 110/113

distance learningEfficient indexing and retrieval methods

2 Decision treesClassification and regression treesRandom Forests

3 Boosting4 Neural networks / Deep LearningMultilayer perceptronsVariations such as convolutional netsExamples and applications

5

Unsupervised methodsClustering

37/37

Overview (Part II: Efros)1 Memory-based/Instance-based learning

k-nearest-neighborProperties of high-dimensional spaces

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 111/113

distance learningEfficient indexing and retrieval methods

2 Decision treesClassification and regression treesRandom Forests

3

Boosting4 Neural networks / Deep LearningMultilayer perceptronsVariations such as convolutional netsExamples and applications

5

Unsupervised methodsClusteringDensity estimation

37/37

Overview (Part II: Efros)1 Memory-based/Instance-based learning

k-nearest-neighborProperties of high-dimensional spaces

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 112/113

distance learningEfficient indexing and retrieval methods2 Decision trees

Classification and regression treesRandom Forests

3

Boosting4 Neural networks / Deep LearningMultilayer perceptronsVariations such as convolutional netsExamples and applications

5

Unsupervised methodsClusteringDensity estimationDimensionality reduction

37/37

Overview (Part II: Efros)1 Memory-based/Instance-based learning

k-nearest-neighborProperties of high-dimensional spaces

8/9/2019 cs189 lecture 1

http://slidepdf.com/reader/full/cs189-lecture-1 113/113

distance learningEfficient indexing and retrieval methods2 Decision trees

Classification and regression treesRandom Forests

3

Boosting4 Neural networks / Deep LearningMultilayer perceptronsVariations such as convolutional netsExamples and applications

5

Unsupervised methodsClusteringDensity estimationDimensionality reductionApplications: Collaborative filtering, etc.

37/37

Recommended