CS194-10 Fall 2011 Introduction to Machine

Содержание

Слайд 2

Course outline

Overview of machine learning (today)
Classical supervised learning
Linear regression,
perceptrons,
neural nets,

Course outline Overview of machine learning (today) Classical supervised learning Linear regression,

SVMs,
decision trees,
nearest neighbors
Learning probabilistic models
Probabilistic classifiers (logistic regression, etc.)
Unsupervised learning, density estimation, EM
Bayes net learning
Time series models

Lecture 1 8/25/11

CS 194-10 Fall 2011, Stuart Russell

Слайд 3

Learning is….

… a computational process for improving performance based on experience

Lecture 1

Learning is…. … a computational process for improving performance based on experience
8/25/11

CS 194-10 Fall 2011, Stuart Russell

Слайд 4

Learning: Why?

Lecture 1 8/25/11

CS 194-10 Fall 2011, Stuart Russell

Learning: Why? Lecture 1 8/25/11 CS 194-10 Fall 2011, Stuart Russell

Слайд 5

Learning: Why?

The baby, rushed by eyes, ears, nose, skin, and insides at

Learning: Why? The baby, rushed by eyes, ears, nose, skin, and insides
once, feels it all as one great blooming, buzzing confusion …
[William James, 1890]

Lecture 1 8/25/11

CS 194-10 Fall 2011, Stuart Russell

Слайд 6

Learning: Why?

The baby, assailed by eyes, ears, nose, skin, and entrails at

Learning: Why? The baby, assailed by eyes, ears, nose, skin, and entrails
once, feels it all as one great blooming, buzzing confusion …
[William James, 1890]

Learning is essential for unknown environments,
i.e., when the designer lacks knowledge

Lecture 1 8/25/11

CS 194-10 Fall 2011, Stuart Russell

Слайд 7

Learning: Why?

Instead of trying to produce a programme to simulate the adult

Learning: Why? Instead of trying to produce a programme to simulate the
mind, why not rather try to produce one which simulates the child's? If this were then subjected to an appropriate course of education one would obtain the adult brain. Presumably the child brain is something like a notebook as one buys it from the stationer's. Rather little mechanism, and lots of blank sheets.
[Alan Turing, 1950]
Learning is useful as a system construction method, i.e., expose the system to reality rather than trying to write it down

Lecture 1 8/25/11

CS 194-10 Fall 2011, Stuart Russell

Слайд 8

Structure of a learning agent

Lecture 1 8/25/11

CS 194-10 Fall 2011, Stuart Russell

Structure of a learning agent Lecture 1 8/25/11 CS 194-10 Fall 2011, Stuart Russell

Слайд 9

Design of learning element

Key questions:
What is the agent design that will implement

Design of learning element Key questions: What is the agent design that
the desired performance?
Improve the performance of what piece of the agent system and how is that piece represented?
What data are available relevant to that piece? (In particular, do we know the right answers?)
What knowledge is already available?

Lecture 1 8/25/11

CS 194-10 Fall 2011, Stuart Russell

Слайд 10

Examples

Supervised learning: correct answers for each training instance
Reinforcement learning: reward sequence, no

Examples Supervised learning: correct answers for each training instance Reinforcement learning: reward
correct answers
Unsupervised learning: “just make sense of the data”

Lecture 1 8/25/11

CS 194-10 Fall 2011, Stuart Russell

Слайд 11

Supervised learning

To learn an unknown target function f
Input: a training set of

Supervised learning To learn an unknown target function f Input: a training
labeled examples (xj,yj) where yj = f(xj)
E.g., xj is an image, f(xj) is the label “giraffe”
E.g., xj is a seismic signal, f(xj) is the label “explosion”
Output: hypothesis h that is “close” to f, i.e., predicts well on unseen examples (“test set”)
Many possible hypothesis families for h
Linear models, logistic regression, neural networks, decision trees, examples (nearest-neighbor), grammars, kernelized separators, etc etc

Lecture 1 8/25/11

CS 194-10 Fall 2011, Stuart Russell

Слайд 12

Supervised learning

To learn an unknown target function f
Input: a training set of

Supervised learning To learn an unknown target function f Input: a training
labeled examples (xj,yj) where yj = f(xj)
E.g., xj is an image, f(xj) is the label “giraffe”
E.g., xj is a seismic signal, f(xj) is the label “explosion”
Output: hypothesis h that is “close” to f, i.e., predicts well on unseen examples (“test set”)
Many possible hypothesis families for h
Linear models, logistic regression, neural networks, decision trees, examples (nearest-neighbor), grammars, kernelized separators, etc etc

Lecture 1 8/25/11

CS 194-10 Fall 2011, Stuart Russell

Слайд 13

Supervised learning

To learn an unknown target function f
Input: a training set of

Supervised learning To learn an unknown target function f Input: a training
labeled examples (xj,yj) where yj = f(xj)
E.g., xj is an image, f(xj) is the label “giraffe”
E.g., xj is a seismic signal, f(xj) is the label “explosion”
Output: hypothesis h that is “close” to f, i.e., predicts well on unseen examples (“test set”)
Many possible hypothesis families for h
Linear models, logistic regression, neural networks, decision trees, examples (nearest-neighbor), grammars, kernelized separators, etc etc

Lecture 1 8/25/11

CS 194-10 Fall 2011, Stuart Russell

Слайд 14

Example: object recognition

x

f(x)

giraffe

giraffe

giraffe

llama

llama

llama

Lecture 1 8/25/11

CS 194-10 Fall 2011, Stuart Russell

Example: object recognition x f(x) giraffe giraffe giraffe llama llama llama Lecture

Слайд 15

Example: object recognition

x

f(x)

giraffe

giraffe

giraffe

llama

llama

llama

X=

f(x)=?

Lecture 1 8/25/11

CS 194-10 Fall 2011, Stuart Russell

Example: object recognition x f(x) giraffe giraffe giraffe llama llama llama X=

Слайд 16

Example: curve fitting

Lecture 1 8/25/11

CS 194-10 Fall 2011, Stuart Russell

Example: curve fitting Lecture 1 8/25/11 CS 194-10 Fall 2011, Stuart Russell

Слайд 17

Example: curve fitting

Lecture 1 8/25/11

CS 194-10 Fall 2011, Stuart Russell

Example: curve fitting Lecture 1 8/25/11 CS 194-10 Fall 2011, Stuart Russell

Слайд 18

Example: curve fitting

Lecture 1 8/25/11

CS 194-10 Fall 2011, Stuart Russell

Example: curve fitting Lecture 1 8/25/11 CS 194-10 Fall 2011, Stuart Russell

Слайд 19

Example: curve fitting

Lecture 1 8/25/11

CS 194-10 Fall 2011, Stuart Russell

Example: curve fitting Lecture 1 8/25/11 CS 194-10 Fall 2011, Stuart Russell

Слайд 20

Learning

data

Learning

knowledge

Lecture 1 8/25/11

CS 194-10 Fall 2011, Stuart Russell

Learning data Learning knowledge Lecture 1 8/25/11 CS 194-10 Fall 2011, Stuart Russell

Слайд 21

Learning

data

Learning

knowledge

prior
knowledge

Lecture 1 8/25/11

CS 194-10 Fall 2011, Stuart Russell

Learning data Learning knowledge prior knowledge Lecture 1 8/25/11 CS 194-10 Fall 2011, Stuart Russell

Слайд 22

Learning

data

Learning

knowledge

prior
knowledge

Lecture 1 8/25/11

CS 194-10 Fall 2011, Stuart Russell

Learning data Learning knowledge prior knowledge Lecture 1 8/25/11 CS 194-10 Fall 2011, Stuart Russell
Имя файла: CS194-10-Fall-2011-Introduction-to-Machine.pptx
Количество просмотров: 41
Количество скачиваний: 0