INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.
-
Upload
brayden-hone -
Category
Documents
-
view
233 -
download
1
Transcript of INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.
![Page 1: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/1.jpg)
INTRODUCTION TO MACHINE LEARNINGDavid KauchakCS 451 – Fall 2013
![Page 2: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/2.jpg)
Why are you here?
What is Machine Learning?
Why are you taking this course?
What topics would you like to see covered?
![Page 3: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/3.jpg)
Machine Learning is…
Machine learning, a branch of artificial intelligence, concerns the construction and study of systems that can learn from data.
![Page 4: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/4.jpg)
Machine Learning is…
Machine learning is programming computers to optimize a performance criterion using example data or past experience.
-- Ethem Alpaydin
The goal of machine learning is to develop methods that can automatically detect patterns in data, and then to use the uncovered patterns to predict future data or other outcomes of interest.
-- Kevin P. Murphy
The field of pattern recognition is concerned with the automatic discovery of regularities in data through the use of computer algorithms and with the use of these regularities to take actions.
-- Christopher M. Bishop
![Page 5: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/5.jpg)
Machine Learning is…
Machine learning is about predicting the future based on the past.
-- Hal Daume III
![Page 6: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/6.jpg)
Machine Learning is…
Machine learning is about predicting the future based on the past.
-- Hal Daume III
TrainingData
learn
model/predictor
past
predict
model/predictor
future
TestingData
![Page 7: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/7.jpg)
Machine Learning, aka
data mining: machine learning applied to “databases”, i.e. collections of data
inference and/or estimation in statistics
pattern recognition in engineering
signal processing in electrical engineering
induction
optimization
![Page 8: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/8.jpg)
Goals of the course: Learn about…Different machine learning problems
Common techniques/tools used theoretical understanding practical implementation
Proper experimentation and evaluation
Dealing with large (huge) data sets Parallelization frameworks Programming tools
![Page 9: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/9.jpg)
Goals of the course
Be able to laugh at these signs(or at least know why one might…)
![Page 10: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/10.jpg)
Administrative
Course page: http://www.cs.middlebury.edu/~dkauchak/classes/cs451/ go/cs451
Assignments Weekly Mostly programming (Java, mostly) Some written/write-up Generally due Friday evenings
Two exams
Late Policy
Honor code
![Page 11: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/11.jpg)
Course expectations
400-level course
Plan to stay busy!
Applied class, so lots of programming
Machine learning involves math
![Page 12: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/12.jpg)
Machine learning problems
What high-level machine learning problems have you seen or heard of before?
![Page 13: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/13.jpg)
Data
examples
Data
![Page 14: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/14.jpg)
Data
examples
Data
![Page 15: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/15.jpg)
Data
examples
Data
![Page 16: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/16.jpg)
Data
examples
Data
![Page 17: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/17.jpg)
Supervised learning
Supervised learning: given labeled examples
label
label1
label3
label4
label5
labeled examples
examples
![Page 18: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/18.jpg)
Supervised learning
Supervised learning: given labeled examples
model/predictor
label
label1
label3
label4
label5
![Page 19: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/19.jpg)
Supervised learning
model/predictor
Supervised learning: learn to predict new example
predicted label
![Page 20: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/20.jpg)
Supervised learning: classification
Supervised learning: given labeled examples
label
apple
apple
banana
banana
Classification: a finite set of labels
![Page 21: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/21.jpg)
Classification Example
Differentiate between low-risk and high-risk customers from their income and savings
![Page 22: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/22.jpg)
Classification Applications
Face recognition
Character recognition
Spam detection
Medical diagnosis: From symptoms to illnesses
Biometrics: Recognition/authentication using physical and/or behavioral characteristics: Face, iris, signature, etc
...
![Page 23: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/23.jpg)
Supervised learning: regression
Supervised learning: given labeled examples
label
-4.5
10.1
3.2
4.3
Regression: label is real-valued
![Page 24: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/24.jpg)
24
Regression Example
Price of a used car
x : car attributes (e.g. mileage)y : price
y = wx+w0
![Page 25: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/25.jpg)
Regression Applications
Economics/Finance: predict the value of a stock
Epidemiology
Car/plane navigation: angle of the steering wheel, acceleration, …
Temporal trends: weather over time
…
![Page 26: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/26.jpg)
Supervised learning: ranking
Supervised learning: given labeled examples
label
1
4
2
3
Ranking: label is a ranking
![Page 27: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/27.jpg)
Ranking example
Given a query anda set of web pages, rank them accordingto relevance
![Page 28: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/28.jpg)
Ranking Applications
User preference, e.g. Netflix “My List” -- movie queue ranking
iTunes
flight search (search in general)
reranking N-best output lists
…
![Page 29: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/29.jpg)
Unsupervised learning
Unupervised learning: given data, i.e. examples, but no labels
![Page 30: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/30.jpg)
Unsupervised learning applicationslearn clusters/groups without any label
customer segmentation (i.e. grouping)
image compression
bioinformatics: learn motifs
…
![Page 31: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/31.jpg)
Reinforcement learning
left, right, straight, left, left, left, straight
left, straight, straight, left, right, straight, straight
GOOD
BAD
left, right, straight, left, left, left, straight
left, straight, straight, left, right, straight, straight
18.5
-3
Given a sequence of examples/states and a reward after completing that sequence, learn to predict the action to take in for an individual example/state
![Page 32: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/32.jpg)
Reinforcement learning example
… WIN!
… LOSE!
Backgammon
Given sequences of moves and whether or not the player won at the end, learn to make good moves
![Page 33: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/33.jpg)
Reinforcement learning example
http://www.youtube.com/watch?v=VCdxqn0fcnE
![Page 34: INTRODUCTION TO MACHINE LEARNING David Kauchak CS 451 – Fall 2013.](https://reader035.fdocuments.net/reader035/viewer/2022081501/56649c945503460f94950611/html5/thumbnails/34.jpg)
Other learning variations
What data is available: Supervised, unsupervised, reinforcement learning semi-supervised, active learning, …
How are we getting the data: online vs. offline learning
Type of model: generative vs. discriminative parametric vs. non-parametric