Location via proxy:   [ UP ]  
[Report a bug]   [Manage cookies]                
SlideShare a Scribd company logo

Neural Networks

•Download as PPT, PDF•
•
N
NikitaRuhela

Artificial neural networks are a form of artificial intelligence inspired by biological neural networks. They are composed of interconnected processing units that can learn patterns from data through training. Neural networks are well-suited for tasks like pattern recognition, classification, and prediction. They learn by example without being explicitly programmed, similarly to how the human brain learns.

1 of 41
ARTIFICIAL NEURAL NETWORKS PRESENTED BY:- Nikita Ruhela B.Tech(I.T.) Vth Sem
KNOWLEDGE-BASED INFORMATION SYSTEMS Knowledge-based system is a program that acquires, represents and uses knowledge for a specific purpose. Consists of a knowledge-base and an inference engine. Knowledge is stored in the knowledge-base while control strategies reside in the separate inference engine.
KNOWLEDGE-BASED INFO SYSTEM knowledge-base inference engine
WHAT ARE NEURAL NETWORKS ? Artificial Neural Network (ANN) :- an information processing paradigm inspired by the HUMAN nervous system. Composed of large number of highly interconnected processing elements (neurons). ANNs, like people, learn by example. An ANN is configured for a specific application, like pattern recognition or data classification, through learning. Learning in biological systems involves synaptic connections between neurons. This is true of ANNs as well.
Why use neural networks ? Knowledge acquisition under noise and uncertainty. Flexible knowledge representation. Efficient knowledge processing. Fault Tolerance . They have learning capability.
Neural networks versus conventional computers ANN Learning approach Not programmed for specific tasks Used in decision making Operation is unpredictable COMPUTERS Algorithmic approach They are necessarily programmed Work on predefined set of instructions Operations are predictable
How does human brain learns ? Brain ,made up of large no. of neurons. Each neuron connects to thousands of neurons, communicates by electrochemical signals. Signals coming are received via SYNAPSES, located at the end of DENDRITES. A neuron sum up the inputs, and if threshold value is reached then it generates a voltage and o/p signal, along the AXON.
FIGURE SHOWING NEURON
SYNAPSE
-:THE ARTIFICIAL NEURON:- Electronically modeled biological neuron. Has many inputs and one output. Has 2 modes -training mode & using mode. Training mode - neuron is trained to fire (or not), for particular input patterns.
-:THE ARTIFICIAL NEURON :- Using mode - when a taught input pattern is detected at input, its associated output becomes current output . If input pattern does not belong in taught list, firing rule is used.
# ARTIFICIAL NEURON #
<((FIRING RULE))> Firing rule calculates whether neuron should fire for an input pattern or not. relates to all the input patterns, seen or unseen. The rule states :- Take collection of training patterns for node, some that cause it to fire (the 1-taught set of patterns) and others which prevent it from firing (the 0-taught set).
<((FIRING RULE))> Then, the patterns not in collection cause node to fire if, they are more similar to patterns in the 1-taught set, than with patterns in the 0-taught set. If there is a tie, then pattern remains in undefined state.
<((FIRING RULE))> Example : a 3-input neuron is taught to output 1 when the input (X1,X2 and X3) is 111 or 101 and to output 0 when the input is 000 or 001. Now, if we give 010,then the neuron will not fire, for 011 o/p is undefined.
~:“PATTERN RECOGNITION”:~ Pattern recognition is implemented by using neural networks. During training, the network is trained to associate outputs with input patterns. The n/w then identifies the input pattern and tries to output associated output pattern.
~:“PATTERN RECOGNITION”:~ The power of neural networks comes to life when a pattern that has no output associated with it, is given as an input. In this case, the network gives the output that corresponds to a taught input pattern that is least different from the given pattern. Example :- recognition of alphabets, symbols etc.
Here the inputs are weighted inputs. Effect of an input on decision making is directly proportional to the weight of that input. Weight is a floating pt. number, can be +ve or –ve. As each input enters the nucleus it is multiplied by its weight . McCulloch And Pitts Model Of Neuron
Neuron then sums these new input values which gives us the activation . If activation is greater than threshold value, the neuron outputs a signal, else zero output. This is typically called a step function . McCulloch And Pitts Model Of Neuron
# ARTIFICIAL NEURON WITH # WEIGHTED INPUTS
In mathematical terms, neuron fires if and only if, X1W1 + X2W2 + X3W3 + ... > T The MCP neuron has the ability to adapt to a particular situation by, changing its weights and/or threshold. Various algorithms exist that cause the neuron to 'adapt'; the most used ones are the Delta rule and the back error propagation. McCulloch And Pitts Model Of Neuron
Architecture Of Neural Networks FEED –FORWARD NETWORKS :- allow signals to travel one way only ; from input to output . no feedback (loops) i.e. the output of any layer does not affect that same layer. Feed-forward ANNs tend to be straight forward networks that associate inputs with outputs. extensively used in pattern recognition .
FEED-FORWARD NETWORK
Architecture Of Neural Networks FEEDBACK NETWORKS :- can have signals traveling in both directions by introducing loops in the network. Feedback networks are dynamic; their 'state' is changing continuously until they reach an equilibrium point. They remain at the equilibrium point until the input changes and a new equilibrium needs to be found . also referred to as interactive or recurrent .
FEEDBACK NETWORK
Architecture Of Neural Networks Network layers :- Artificial neural network mostly consists of three groups, or layers, Input Layer - activity of input units represents raw information that is fed into the network. Hidden Layer - activity of each hidden unit is determined by activities of input units and weights on connections between input and hidden units. Output Layer - behavior of the output units depends on the activity of the hidden units and the weights between the hidden and output units.
Layers Of Neural Network
Architecture Of Neural Networks Perceptrons :- It is an MCP model with some additional, fixed, pre-processing . Units A1, A2, Aj , Ap are called association units and they extract specific, localized features from input images. mimic the basic idea behind the human visual system.
Most common methods used are :- Supervised Learning Unsupervised Learning Learning Methods
Incorporates an external teacher Each output unit is told what its desired response to input signals should be. During learning process global information is required. Supervised learning includes error- correction and reinforcement learning. Supervised Learning
There is problem of error convergence i.e. minimization of error between desired and computed values . The aim is to determine a set of weights which minimizes the error. A well-known method is least mean square (LMS) convergence. Supervised Learning
Uses no external teacher, based upon only local information. Also referred to as self-organizing, in sense that it self-organizes data presented to network, detects their emergent collective properties. Methods of unsupervised learning are Hebbian and competitive learning . Unsupervised Learning
Learning consists of two phases- training phase and operation phase. We say that neural network learns off-line if learning phase and operation phase are distinct. A neural network learns on-line if it learns and operates at the same time. Usually, supervised learning is performed off-line, whereas unsupervised learning is performed on-line. Learning Contd....
Character Recognition Since neural networks are best at identifying patterns or trends in data, they are well suited for prediction or forecasting needs including: sales forecasting industrial process control data validation risk management Applications
Neural networks are also used for Genetic pattern recognition Drug discovery Flow Cytometric Analysis of Leukemia Also used in field of Robotics , Facial Animation, Lip Reading , Event Prediction and many more fields. Applications
Learning mode Prediction mode Applications
Shape of Lips Lip reading Applications
The computing world has a lot to gain from neural networks. Their ability to learn by example makes them very flexible and powerful. Further, there is no need to devise an algorithm in order to perform a specific task; i.e. there is no need to understand the internal mechanisms of that task. They are also very well suited for real time systems because of their fast response and computational times which are due to their parallel architecture. Conclusion
Neural networks also contribute to other areas of research such as neurology and psychology. Finally, I would like to state that even though neural networks have a huge potential we will only get the best of them when they are integrated with computing, AI and related subjects. Conclusion
en.wikipedia.org www.sciencedirect.com www.learnartificialneuralnetworks.com www.doc.ic.ac.uk www.statsoft.com NEURAL NETWORKS IN COMPUTER INTELLIGENCE ,by – LiMin Fu REFERENCES
Thank You !

More Related Content

Neural Networks

  • 1. ARTIFICIAL NEURAL NETWORKS PRESENTED BY:- Nikita Ruhela B.Tech(I.T.) Vth Sem
  • 2. KNOWLEDGE-BASED INFORMATION SYSTEMS Knowledge-based system is a program that acquires, represents and uses knowledge for a specific purpose. Consists of a knowledge-base and an inference engine. Knowledge is stored in the knowledge-base while control strategies reside in the separate inference engine.
  • 3. KNOWLEDGE-BASED INFO SYSTEM knowledge-base inference engine
  • 4. WHAT ARE NEURAL NETWORKS ? Artificial Neural Network (ANN) :- an information processing paradigm inspired by the HUMAN nervous system. Composed of large number of highly interconnected processing elements (neurons). ANNs, like people, learn by example. An ANN is configured for a specific application, like pattern recognition or data classification, through learning. Learning in biological systems involves synaptic connections between neurons. This is true of ANNs as well.
  • 5. Why use neural networks ? Knowledge acquisition under noise and uncertainty. Flexible knowledge representation. Efficient knowledge processing. Fault Tolerance . They have learning capability.
  • 6. Neural networks versus conventional computers ANN Learning approach Not programmed for specific tasks Used in decision making Operation is unpredictable COMPUTERS Algorithmic approach They are necessarily programmed Work on predefined set of instructions Operations are predictable
  • 7. How does human brain learns ? Brain ,made up of large no. of neurons. Each neuron connects to thousands of neurons, communicates by electrochemical signals. Signals coming are received via SYNAPSES, located at the end of DENDRITES. A neuron sum up the inputs, and if threshold value is reached then it generates a voltage and o/p signal, along the AXON.
  • 10. -:THE ARTIFICIAL NEURON:- Electronically modeled biological neuron. Has many inputs and one output. Has 2 modes -training mode & using mode. Training mode - neuron is trained to fire (or not), for particular input patterns.
  • 11. -:THE ARTIFICIAL NEURON :- Using mode - when a taught input pattern is detected at input, its associated output becomes current output . If input pattern does not belong in taught list, firing rule is used.
  • 13. <((FIRING RULE))> Firing rule calculates whether neuron should fire for an input pattern or not. relates to all the input patterns, seen or unseen. The rule states :- Take collection of training patterns for node, some that cause it to fire (the 1-taught set of patterns) and others which prevent it from firing (the 0-taught set).
  • 14. <((FIRING RULE))> Then, the patterns not in collection cause node to fire if, they are more similar to patterns in the 1-taught set, than with patterns in the 0-taught set. If there is a tie, then pattern remains in undefined state.
  • 15. <((FIRING RULE))> Example : a 3-input neuron is taught to output 1 when the input (X1,X2 and X3) is 111 or 101 and to output 0 when the input is 000 or 001. Now, if we give 010,then the neuron will not fire, for 011 o/p is undefined.
  • 16. ~:“PATTERN RECOGNITION”:~ Pattern recognition is implemented by using neural networks. During training, the network is trained to associate outputs with input patterns. The n/w then identifies the input pattern and tries to output associated output pattern.
  • 17. ~:“PATTERN RECOGNITION”:~ The power of neural networks comes to life when a pattern that has no output associated with it, is given as an input. In this case, the network gives the output that corresponds to a taught input pattern that is least different from the given pattern. Example :- recognition of alphabets, symbols etc.
  • 18. Here the inputs are weighted inputs. Effect of an input on decision making is directly proportional to the weight of that input. Weight is a floating pt. number, can be +ve or –ve. As each input enters the nucleus it is multiplied by its weight . McCulloch And Pitts Model Of Neuron
  • 19. Neuron then sums these new input values which gives us the activation . If activation is greater than threshold value, the neuron outputs a signal, else zero output. This is typically called a step function . McCulloch And Pitts Model Of Neuron
  • 20. # ARTIFICIAL NEURON WITH # WEIGHTED INPUTS
  • 21. In mathematical terms, neuron fires if and only if, X1W1 + X2W2 + X3W3 + ... > T The MCP neuron has the ability to adapt to a particular situation by, changing its weights and/or threshold. Various algorithms exist that cause the neuron to 'adapt'; the most used ones are the Delta rule and the back error propagation. McCulloch And Pitts Model Of Neuron
  • 22. Architecture Of Neural Networks FEED –FORWARD NETWORKS :- allow signals to travel one way only ; from input to output . no feedback (loops) i.e. the output of any layer does not affect that same layer. Feed-forward ANNs tend to be straight forward networks that associate inputs with outputs. extensively used in pattern recognition .
  • 24. Architecture Of Neural Networks FEEDBACK NETWORKS :- can have signals traveling in both directions by introducing loops in the network. Feedback networks are dynamic; their 'state' is changing continuously until they reach an equilibrium point. They remain at the equilibrium point until the input changes and a new equilibrium needs to be found . also referred to as interactive or recurrent .
  • 26. Architecture Of Neural Networks Network layers :- Artificial neural network mostly consists of three groups, or layers, Input Layer - activity of input units represents raw information that is fed into the network. Hidden Layer - activity of each hidden unit is determined by activities of input units and weights on connections between input and hidden units. Output Layer - behavior of the output units depends on the activity of the hidden units and the weights between the hidden and output units.
  • 27. Layers Of Neural Network
  • 28. Architecture Of Neural Networks Perceptrons :- It is an MCP model with some additional, fixed, pre-processing . Units A1, A2, Aj , Ap are called association units and they extract specific, localized features from input images. mimic the basic idea behind the human visual system.
  • 29. Most common methods used are :- Supervised Learning Unsupervised Learning Learning Methods
  • 30. Incorporates an external teacher Each output unit is told what its desired response to input signals should be. During learning process global information is required. Supervised learning includes error- correction and reinforcement learning. Supervised Learning
  • 31. There is problem of error convergence i.e. minimization of error between desired and computed values . The aim is to determine a set of weights which minimizes the error. A well-known method is least mean square (LMS) convergence. Supervised Learning
  • 32. Uses no external teacher, based upon only local information. Also referred to as self-organizing, in sense that it self-organizes data presented to network, detects their emergent collective properties. Methods of unsupervised learning are Hebbian and competitive learning . Unsupervised Learning
  • 33. Learning consists of two phases- training phase and operation phase. We say that neural network learns off-line if learning phase and operation phase are distinct. A neural network learns on-line if it learns and operates at the same time. Usually, supervised learning is performed off-line, whereas unsupervised learning is performed on-line. Learning Contd....
  • 34. Character Recognition Since neural networks are best at identifying patterns or trends in data, they are well suited for prediction or forecasting needs including: sales forecasting industrial process control data validation risk management Applications
  • 35. Neural networks are also used for Genetic pattern recognition Drug discovery Flow Cytometric Analysis of Leukemia Also used in field of Robotics , Facial Animation, Lip Reading , Event Prediction and many more fields. Applications
  • 36. Learning mode Prediction mode Applications
  • 37. Shape of Lips Lip reading Applications
  • 38. The computing world has a lot to gain from neural networks. Their ability to learn by example makes them very flexible and powerful. Further, there is no need to devise an algorithm in order to perform a specific task; i.e. there is no need to understand the internal mechanisms of that task. They are also very well suited for real time systems because of their fast response and computational times which are due to their parallel architecture. Conclusion
  • 39. Neural networks also contribute to other areas of research such as neurology and psychology. Finally, I would like to state that even though neural networks have a huge potential we will only get the best of them when they are integrated with computing, AI and related subjects. Conclusion
  • 40. en.wikipedia.org www.sciencedirect.com www.learnartificialneuralnetworks.com www.doc.ic.ac.uk www.statsoft.com NEURAL NETWORKS IN COMPUTER INTELLIGENCE ,by – LiMin Fu REFERENCES