Title: Neural Networks Introduction and Review
1Neural Networks- Introduction and Review -
2History of Neural Networks
- Attempts to mimic the human brain date back to
work in the 1930s, 1940s, 1950s by Alan Turing,
Warren McCullough, Walter Pitts, Donald Hebb and
James von Neumann - 1943 McCulloch-Pitts neuron as comp. elem
- 1948 Wiener cybernatics
- 1949 Hebb learning rule
- 1957 Rosenblatt at Cornell developed Perceptron,
a hardware neural net for character recognition - 1959 Widrow and Hoff at Stanford developed
Adaline for adaptive control of noise on
telephone lines - 1960 Widrow-Hoff least mean square algorithm
3History of Neural Networks
- Recession
- 1969 Minsky-Papert limitations perceptron model
- Linear Separability in Perceptrons
4History of Neural Networks
- Revival, mathematically tied together many of the
ideas from previous research. - 1982 Hopfield recurrent network model
- 1982 Kohonen self-organizing maps
- 1986 Rumelhart et. al. backpropagation
- universial approximation
- Since then, growth has exploded. Over 80 of
Fortune 500 have neural net RD programs. - Thousands of research papers
- Commercial software applications
5Application with neural network
- Forecasting/Market Prediction finance and
banking - Manufacturing quality control, fault diagnosis
- Medicine analysis of electrocardiogram data,
RNA DNA sequencing, drug development without
animal testing - Pattern/Image recognition handwriting
recognition, airport bomb detection - Optimization without Simplex
- Control process, robotics
6The Biological Neuron
Neurons are brain cells, it is estimated that
there are 1012 neurons and 1014 synaptic
connections in the human brain
7Biological Neurons
dendrites
axon
synapses
The information transmission happens at the
synapses.
8Neural Dynamics
Action potential 100mV Activation threshold
20-30mV Rest potential -65mV Spike time
1-2ms Refractory time 10-20ms
Action potential
Refractory time
9Key to intelligence
- Synapse weight adjustment.
- Connection strength
- Each neuron receives input from nearly 50,000 to
80,000 other neurons in the human brain. - The contribution of the signals depends on the
strength of the synaptic connection.
10Simple Neuron
- Nodes have input signals. Dendrites carry an
impulse to the neuron - Nodes have one output signal. Axons carry signal
out of neuron and synapses are local regions
where signals are transmitted from the axon of
one neuron to dendrites of another - Input signal weights are summed at each node.
Nerve impulses are binary they are go or no
go. Neurons sum up the incoming signal and fire
if a threshold value is reached.
11Artificial neurons
Neurons work by processing information. They
receive and provide information in form of spikes.
x1 x2 x3 xn-1 xn
w1
Output
w2
Inputs
y
w3
.
.
.
wn-1
wn
The McCullogh-Pitts model
12Binary Neurons
Stimulus
Response
on
Hard threshold
off
- ex Perceptrons, Hopfield NNs, Boltzmann Machines
- Main drawbacks can only map binary functions,
biologically implausible.
13Analog Neurons
Stimulus
Response
on
Soft threshold
off
- ex MLPs, Recurrent NNs, RBF NNs...
- Main drawbacks difficult to process time
patterns, biologically implausible.
14Spiking Neurons
Stimulus
- spike and afterspike potential
- urest resting potential
- e(t,u(t)) trace at time t of input at time t
- threshold
- xj(t) output of neuron j at time t
- wij efficacy of synapse from neuron i to neuron
j - u(t) input stimulus at time t
Response
Firing pattern interpretation (spike trains of
units) Timing of spike trains encodes
information (time to first spike, phase of
signal, correlation and synchronicity
15Spiking Neuron Dynamics
16Artificial Neural Networks
Output layer
Hidden layers
fully connected
Input layer
sparsely connected
17Feedforward ANN Architectures
- Information flow unidirectional
- Static mapping yf(x)
- Multi-Layer Perceptron (MLP)
- Radial Basis Function (RBF)
- Kohonen Self-Organising Map (SOM)
18Recurrent ANN Architectures
- Feedback connections
- Dynamic memory y(t1)f(x(t),y(t),s(t))
t?(t,t-1,...) - Jordan/Elman ANNs
- Hopfield
- Adaptive Resonance Theory (ART)
19Activation functions
Linear
Sigmoid
Hyperbolic tangent
20Neural network mathematics
Inputs
a
Output
y
f
b
MLP neural network
Gaussian RBF
a
f
y
b
f
21ANN Capabilities
- Learning
- Approximate reasoning
- Generalisation capability
- Noise filtering
- Parallel processing
- Distributed knowledge base
- Fault tolerance
22Properties of Neural Networks
- Supervised networks are universal approximators
- Theorem Any limited function can be
approximated by a neural network with a finite
number of hidden neurons to an arbitrary
precision - Type of Approximators
- Linear approximators for a given precision, the
number of parameters grows exponentially with the
number of variables (polynomials) - Non-linear approximators (NN), the number of
parameters grows linearly with the number of
variables - Knowledge base not transparent (black box)
(Partially resolved) - Learning sometimes difficult/slow
- Limited storage capability
23Learning in biological systems as optimisation
- Learning learning by adaptation
- The young animal learns that the green fruits are
sour, while the yellowish/reddish ones are sweet.
The learning happens by adapting the fruit
picking behavior. The animal likes to eat many
energy rich, juicy fruits that make its stomach
full, and makes it feel happy. - At the neural level the learning happens by
changing of the synaptic strengths, eliminating
some synapses, and building new ones. the
objective of learning in biological organisms is
to optimise the amount of available resources,
happiness, or in general to achieve a closer to
optimal state
24Learning principle for artificial neural networks
- Maintaining synaptic strength needs energy, it
should be maintained at those places where it is
needed, and it shouldnt be maintained at places
where its not needed - ENERGY MINIMIZATION
- We need an appropriate definition of energy for
artificial neural networks, and having that we
can use mathematical optimisation techniques to
find how to change the weights of the synaptic
connections between neurons. - ENERGY measure of task performance error
25Neural Networks Learning
- Supervised learning
- Classification
- Control
- Function approximation
- Associative memory
- Unsupervised learning
- Clustering
- Reinforcement learning
- Control
26Unsupervised Learning
- ANN adapts weights to cluster input data
- Hebbian learning
- Connection stimulus-response strengthened
(hebbian) - Competitive learning algorithms
- Kohonen ART
- Input weights adjusted to resemble stimulus
27Hebbian Learning
General Formulation
Kohonen, Competitive (ART)
Hebbian
llearning coefficient wijconnection from neuron
xj to yi
- Hebb postulate (1948)
- Correlation-based learning
- Connections between concurrently firing neurons
are strengthened - Experimentally verified (1973)
28Supervised Learning
- Teacher presents ANN input-output pairs
- ANN weights adjusted according to error
- Iterative algorithms (e.g. Delta rule, BP rule)
- One-shot learning (Hopfield)
- Quality of training examples is critical
29Delta Rule
- Least Mean Squares
- Widrow-Hoff iterative delta rule
- Gradient descent of the error surface
- Guaranteed to find minimum error configuration in
single layer ANNs
30Gradient Learning
is the learning parameter (usually a constant)
31Learning with a perceptron
A perceptron is able to learn a linear function.
Perceptron
Data
Error
Learning
32Learning with RBF neural networks
An RBF neural network learns a nonlinear function.
RBF neural network
Data
Error
Learning
33Learning with general optimization
MLP neural network with a single hidden layer
34Learning with general optimization
35Reinforcement Learning
- Sequential tasks
- Desired action may not be known
- Critic evaluation of ANN behaviour
- Weights adjusted according to critic
- May require credit assignment
- Population-based learning
- Evolutionary Algorithms
- Swarming Techniques
- Immune Networks
36Learning Summary
Artificial Neural Networks
Unsupervised
Supervised
Unsupervised
Supervised
Elman, Jordan, Hopfield
MLP, RBF
ART
Kohonen, Hebbian