1. Neural Nets Applications
Introduction
Spring 2017
Instructor: Tai-Yue (Jason) Wang
Department of Industrial and Information Management
Institute of Information Management

2. Outline(1/2) What is a Neural Network? Benefit of Neural Networks
Structural Levels of Organization in the Brain
Models of a Neuron
Network Architectures
Artificial Intelligence and Neural Networks

3. Outline(2/2) Existing Applications Possible Applications Experiment I
Experiment II
Other names for Neural Networks
Who are the key player?

4. What is a Neural Networks(1/5)
Neural networks technology is not trying to produce biological machine
but is trying to mimic nature’s approach in order to capture some of nature’s capabilities.

5. What is a Neural Networks (2/5)
Definition:
A neural network is a massively parallel distributed processor that has a natural propensity for storing experiential knowledge and making it available for use.

6. What is a Neural Networks (3/5)
It resembles the brain in two respects:
Knowledge is acquired by the network through a learning process.
Interneuron connection strengths known as synaptic weight are used to store the knowledge.

7. What is a Neural Networks (4/5)
The Human Brain:
Five to six orders of magnitude slower than silicon logic gates
With 60 trillion synapses or connections
A highly complex, nonlinear, and parallel computer.
Figure 1.1

8. What is a Neural Networks (5/5)

9. Benefits of Neural Networks (1/2)
Nonlinearity
Input-Output Mapping
Adaptivity
Evidential Response
Contextual Information

10. Benefits of Neural Networks (2/2)
Fault Tolerance
Implementability
Uniformity of Analysis and Design
Neurobiological Analogy

11. Structural Levels of Organization in the Brain (1/3)
Figure 1.2
Figure 1.3

12. Structural Levels of Organization in the Brain (2/3)

13. Structural Levels of Organization in the Brain (3/3)

14. Models of a Neuron (1/6) Figure 1.4
Three basic elements of the neuron model:
A set of synapses or connecting links, each of which is characterized by a weight or strength of its own.
An adder for summing the input signals, weighted by the respective synapses of the neuron; the operations described here constitute a linear combiner.
An activation function for limiting the amplitude of the output of a neuron.

15. Models of a Neuron (2/6)

16. Models of a Neuron (3/6) Mathematical terms: where: xj: input signals
wkj: synaptic weights
uk: linear combiner output
θk:: threshold
f() : activation function
yk: output signal

17. Models of a Neuron (4/6) Types of activation function:
a. Threshold function

18. Models of a Neuron (5/6) Types of activation function:
b. Piecewise-linear function

19. Models of a Neuron (6/6) Types of activation function:
c. Sigmoid Function

20. Network Architecture (1/5)
single-layer feedforward network

21. Network Architecture (2/5)
Multilayer feedforward network (fully connected

22. Network Architecture (3/5)
Multilayer feedforward network (partially connected

23. Network Architecture (4/5)
Recurrent networks

24. Network Architecture (5/5)
Lattice Structures

25. Artificial Intelligence and Neural Networks (1/5)
AI system

26. Artificial Intelligence and Neural Networks (2/5)
a. Representation
- use a language of symbol structures to represent both general knowledge about a problem domain of interest and specific knowledge about the solution to the problem.

27. Artificial Intelligence and Neural Networks (3/5)
b. Reasoning
- the ability to solve the problems
- be able to express and solve a broad range of problems and problem types.
- be able to make explicit and inplicit information known to it
- have a control mechanism that determines which operations to apply to a particular problem.

28. Artificial Intelligence and Neural Networks (4/5)
c. Learning
- Fig 1.27
- Inductive, rules are from raw data and experience
- Deductive, rules are used to determine specific facts

29. Artificial Intelligence and Neural Networks (5/5)

30. Existing Applications(1/4)
Long distance echo adaptive fitter adaptive noise canceling
-- ADALINE
Mortgage risk evaluator
Bomb sniffer
-- SNOOPE
-- JFK airport

31. Existing Applications(2/4)
4. Process Monitor -- GTE used in fluorescent bulb plant.-- To determine optimum manufacturing condition. -- To indicate what controls need to be adjusted , and potentially to even shut down the line. -- Statistics could provide same result but with huge data.

32. Existing Applications(3/4)
5. Word Recognizer
--Intel used single speaker on limited vocabulary.
6. Blower Motor Checker
--Siemens used to check Blower motor noise is heater.
7. Medical events

33. Existing Applications(4/4)
8. US postal office for hand written 9. Airline marketing tactician.

34. Possible Applications(1/6)
1. Biological --Learning more about the brain and other systems --Modeling retina , cochlea 2. Environmental --Analyzing trends and patterns --Forecasting weather

35. Possible Applications(2/6)
3. Business
--Evaluating probability of oil in geological formations
--Identifying corporate candidates for special positions
--Mining corporate databases
--Optimizing airline seating and fee schedules
--Recognizing handwritten characters, such as Kanji

36. Possible Applications(3/6)
4. Financial --Assessing credit risk --Identifying forgeries --Interpreting handwritten forms --Rating investments and analyzing portfolios

37. Possible Applications(4/6)
5. Manufacturing
--Automating robots and control system (with machine vision and sensors for pressure. temperature, gas, etc.)
--Controlling production line processes
--Inspecting for quality
--Selecting parts on an assembly line

38. Possible Applications(5/6)
6. Medical
--Analyzing speech in hearing aids for the profoundly deaf
--Diagnosing/prescribing treatments from symptoms
--Monitoring surgery
--Predicting adverse drug reactions
--Reading X-rays
--Understanding cause of epileptic seizures

39. Possible Applications(6/6)
7. Military
--Classifying radar signals
--Creating smart weapons
--Doing reconnaissance
--Optimizing use of scarce resources
--Recognizing and tracking targets

40. Experiment I
to understand a sentence are character a time is much larger than one word a time
conventional computer processes its input one of a time, working sequentially
our eyes look at the whole sentence
vowels are missing
three different groupings

41. Experiment II(1/2) Toss a chalk to another one
-- it is hard in dynamics estimate the speed, the trajectory, the weight in real time
-- computer must be faster

42. Experiment II(2/2) But -- our brain is slower than computer
-- messages in the brain can travel at speeds up to 268 miles/hour
--computer is million times faster
-- our brain still better than computer
Why?
 parallel processing

43. Other Names for Artificial Neural Networks
Parallel/distributed processing models
Connectivist/connectionism models
adaptive systems
self-organizing systems
Neurocomputing
Neuromorphic systems
Self-learning systems

44. Who Are the Key Players? (1/2)
1. Medical and theoretical neurobiologists --Neurophysiology, drug chemistry , molecular biology 2. Computer and information scientists --Information theory 3. Adaptive control theorists/psychologists --Merging learning and control theory

45. Who Are the Key Players? (2/2)
4. Adaptive systems -- researchers/biologists --Self-organization of living species 5. AI researchers --Machine learning mechanisms