machine learning. learning agent any other agent

24
Machine Learning

Upload: zoe-newton

Post on 27-Dec-2015

221 views

Category:

Documents


2 download

TRANSCRIPT

Page 1: Machine Learning. Learning agent Any other agent

Machine Learning

Page 2: Machine Learning. Learning agent Any other agent

Learning agent

Any other agent

Page 3: Machine Learning. Learning agent Any other agent

Learning agent

learning element: responsible for making improvementsperformance element : responsible for selecting external actionscritic : gives feedback to the learning element on how the agent is doing with respect to a fixed performance standard and determines how the performance element should be modified to do better in the future.problem generator : responsible for suggesting actions that will lead to new and informative experiences

Page 4: Machine Learning. Learning agent Any other agent

Structural organization of levels in biological nervous systems.

Page 5: Machine Learning. Learning agent Any other agent

Human brain

Page 6: Machine Learning. Learning agent Any other agent

Artificial neurons

Neurons work by processing information.

The McCullogh-Pitts model

Inputs

Outputw2

w1

w3

wn

wn-1

. . .

x1

x2

x3

xn-1

xn

y)(;

1

zHyxwzn

iii

Page 7: Machine Learning. Learning agent Any other agent

Artificial neural network (ANN)is a mathematical model or computational model based on biological neural networks Artificial Neural Network consists of neurons arranged in layers Neurons act as parallel processorNeurons are connected with each other vi connection.there are weights associated with connectionsImplementation:

Learningtesting

Page 8: Machine Learning. Learning agent Any other agent

Artificial neural networks

Inputs

Output

An artificial neural network is composed of many artificial neurons that are linked together according to a specific network architecture. The objective of the neural network is to transform the inputs into meaningful outputs.

Page 9: Machine Learning. Learning agent Any other agent

Artificial neural networks

Dendrites: Input Layer

Axon : Output Layer

Soma: Net( weighted sum of input y_in) and activation function

Synapse: Weights

Page 10: Machine Learning. Learning agent Any other agent

Why use ANN?-Adaptive learning: An ability to learn how to do tasks based on

the data given for training or initial experience.

-Self-Organization: An ANN can create its own organization or representation of the information it receives during learning

time.

-Real Time Operation: ANN computations may be carried out in parallel, and special hardware devices are being designed and

manufactured which take advantage of this capability.

-Fault Tolerance via Redundant Information Coding: Partial destruction of a network leads to the corresponding degradation

of performance. However, some network capabilities may be retained even with major network damage.

Page 11: Machine Learning. Learning agent Any other agent

ANN Characterization

ANN can be characterized by:Activation function

Weights Adjustment (learning algorithm)

Architecture

Page 12: Machine Learning. Learning agent Any other agent

Learning Algorithm

Learning in ANN is Weights adjustment to get desired output

To minimize the error

To gain more experience

LearningSupervised

unsupervised

Page 13: Machine Learning. Learning agent Any other agent

Supervised Learning

There is supervisor during learning processInput and output are knownThe job of ANN is to classify any new input according to known classesExample : teaching baby the difference pens and other thingsLVQ (learning vector quantization)

Page 14: Machine Learning. Learning agent Any other agent

Unsupervised learning

Input known but output unknown

The classes are unknown to ANN

Job of ANN is to find similarities between input and divide them into categories (cluster)

SOM (Self organizing map)

Page 15: Machine Learning. Learning agent Any other agent

Architecture

Feed forwardallow signals to travel one way only; from input to output.

There is no feedback (loops)

Multi layer

Page 16: Machine Learning. Learning agent Any other agent

Architecture

Feedback networks signals travelling in both directions by introducing loops in the network

Page 17: Machine Learning. Learning agent Any other agent

Decision Tree Representation

Outlook

Humidity Wind

SunnyOvercast

Rain

High Normal Strong Weak

Decision Tree for the concept PlayTennis

Page 18: Machine Learning. Learning agent Any other agent

Pattern recognition system.

Page 19: Machine Learning. Learning agent Any other agent

Flow chart of machine learning for pattern recognition.

Page 20: Machine Learning. Learning agent Any other agent

Bayes classifier

Page 21: Machine Learning. Learning agent Any other agent

Approaches

Page 22: Machine Learning. Learning agent Any other agent

What is a Concept?

A Concept is a a subset of objects or events defined over a larger set [Example: The concept of a bird is the subset of all objects (i.e., the set of all things or all animals) that belong to the category of bird.]

Things

Animals

Birds

Cars

Page 23: Machine Learning. Learning agent Any other agent

What is Concept-Learning?

Given a set of examples labeled as members or non-members of a concept, concept-learning consists of automatically inferring the general definition of this concept.

In other words, concept-learning consists of approximating a boolean-valued function from training examples of its input and output.

Page 24: Machine Learning. Learning agent Any other agent