You are on page 1of 7

.

Research Methodology
M.Tech Semester III (VLSI Design)
Assignment No.1
Submitted by:
Anu Prema Rajendran
Roll no: 02

Q1.Artificial Neural networks (ANNs) have been used as a popular


research and implementation paradigm in multiple domains .Justify

Ans –1)Artificial neural network (ANN) modeling methods are becoming more widely used as
both a research and application paradigm across a much wider variety of business, medical,
engineering, and social science disciplines.
2)Artificial neural networks are increasingly popular in today's business fields. They have been
hailed as the greatest technological advance since the invention of transistors. Recently,
applications of artificial neural networks have been increasing in business. More and more
development tools have emerged on the market. Many neural-net systems have been shown to
work well in identifying intricate patterns, learning from experience, reaching some conclusions,
and making predictions.
3)According to J. Clarke Smith, executive vice president of Sears Mortgage Corporation, neural-
net systems have already been at work for over 10 years in the finance world. Today, they are
widely applicable to risk management and forecasting . Since the various neural-net systems now
in use are implemented with mathematically sound principles, they hold out promise for future
application.
4)They are feasible for those business applications which require the solution of very complex
system equations, recognizing patterns from imperfect inputs, and adapting decisions to
changing environment.
5)The combination or triangulation of ANN methods with more traditional methods can
facilitate the development of high-quality research models and also improve output performance
for real world applications.

Q2.What are the characteristics of the problem suitable for ANN ?


Ans-ANN seems to provide solutions for the problems characterized by some
or all of the following: High dimensional problem space,Complex ,unknown
interactions between problem variables.Solution space that may be empty or
contain a number of useful solutions

Q3. What are the advantages and disadvantages of ANN in relation to


computational burden, proneness to overfitting, and the empirical
nature of model development.?
Ans- Neural networks offer a number of advantages, including requiring less formal statistical
training, ability to implicitly detect complex nonlinear relationships between dependent and
independent variables, ability to detect all possible interactions between predictor variables, and
the availability of multiple training algorithms.
Disadvantages include its "black box" nature i.e., it is difficult to evaluate the contribution of the
independent variables, greater computational burden, proneness to overfitting, and the empirical
nature of model development . computational burden disadvantage:
the computational complexity depends on the dimensionality of the input space. ANNs can suffer
from multiple local minima, proneness to overfitting:Methods for controlling the bias/variance
tradeoff typically assume that overfitting or overtraining is a global phenomenon. For multi-layer
perceptron (MLP) neural networks, global parameters such as the training time (e.g. based on
validation tests), network size, or the amount of weight decay are commonly used to control the
bias/variance tradeoff. However, the degree of overfitting can vary significantly throughout the
input space of the model. We show that overselection of the degrees of freedom for anMLP
trained with backpropagation can improve the approximation in regions of underfitting, while
not significantly overfitting in other regions. This can be a significant advantage over other
models. Furthermore, we show that “better” learning algorithms such as conjugate gradient can
in fact lead to worse generalization, because they can be more prone to creating varying degrees
of overfitting in different regions of the input space.

Q4. Compare the traditional computing techniques with ANN


techniques.
Ans-a)Traditional computing methods work well for problems that can be
well characterized.
b)As the complexity of the system increases, these methods simply
demand too much computing resources and become too slow.
c)Neural networks are structured to provide the capability to solve
problems without the benefits of an expert and without the need of
programming.
d)Traditional processors lack both the processing power and a step-by-
step methodology.
e)The biggest demand of ANN is that the process is not simply logic. It
involves an empirical skill, an intuitive feel as to how a network might be
created.

Q5. State the need for training the neural network.


Ans-
After initializing weight vectors, the key features whose modulation types
have been known are sent to the network.
The training is continued until the weight vectors have been stabilized.
During testing, the actual output vector is not always as the same as the
expected vector because of the noise. The type with the nearest distance is
chosen.
Q7. What are the steps involved in the design of neurocontroller. What
are the advantages of neurocontrol ?
Ans-Neural control:
is directly applicable to non-linear systems because of their ability to map
any arbitrary transfer function
has a parallel structure thereby permitting high computational speeds.
has a much higher reliability and fault tolerance than conventional
controllers
Neural controllers can be trained from prior operational data and can
generalize when subjected to causes that they were not trained with
Neural controllers have the inherent ability to process multiple inputs and
generate multiple outputs simultaneously, making them ideal for
multivariable intelligent control.

Steps involved in the design of neurocontroller

1)System Identification
Neural network model of the plant that is to be controlled is developed.
2)Control Design
The neural network plant model to design (or train) the controller is used.

The popular neural network architectures for prediction and control that
have implemented in neural networks are
Model predictive control
Feedback Linearization control
Model reference control

Model Predictive control


System Identification
The neural network predictive controller uses a neural network model of a
nonlinear plant to predict future plant performance.
The controller then calculates the control input that will optimize plant
performance over a specified future time horizon.
The neural network plant model uses previous inputs and previous plant
outputs to predict future values of the plant output.
This network can be trained offline in batch mode, using data collected from
the operation of the plant.

Predictive control
The predictions are used by a numerical optimization program to determine
the control signal that minimizes the following performance criterion over
the specified horizon:
U
Feedback Linearization Control
Transforms nonlinear system dynamics into linear dynamics by canceling the
nonlinearities.
The problem with using this controller is that if we want to train a neural
network to create the function that will minimize mean square error, we
need to use dynamic back -propagation .
This can be quite slow.

Model Reference Control


It uses two neural networks: a controller network and a plant model network
The plant model is identified first and then the controller is trained so that
the plant output follows the reference model output.
Each network has two layers, and we can select the number of neurons to
use in the hidden layers.

8. Discuss the application of ANN in image processing. Give relevant


examples.
Ans- Artificial neural networks (ANNs) can overcome some of these
difficulties by interpreting images quickly and effectively. ANNs are
composed of numerous processing elements (PEs) arranged in various
layers, with interconnections between pairs of PEs(Haykin 1994;
Kartalopoulos 1996; Kasabov 1996). They are designed to emulate the
structure of natural neural networks such as those of a human brain. For
most ANNs, PEs in each layer are fully connected with PEs in the adjacent
layer or layers, but are not connected to other PEs in the same layer. The
PEs simulate the function of the neurons in natural neural networks, while
the interconnections between them mimic the functions of dendrites and
axons. There have been many applications of ANNs reported for the
interpretation of images in the agri-food industry.There have been many
applications of ANNs reported for the interpretation of images in the agri-
food industry.

Q9. Discuss the efficacy of utilizing nonconcurrent triangulation of


method with ANNs, specifically the preselection of variables with
recognized statistical techniques.

ANS-ANNs are black boxes (i.e., it is difficult to evaluate the contribution of the independent
variables) .Their black box nature has led to adoption resistance by numerous-business
related disciplines . Methodological triangulation may help to overcome these adoption and
usage reservations as well as providing a means for improving the overall efficacy of ANN
applications. Methodological triangulation is the utilization of multiple methods on the same
problem (empirical data) to gain confidence in the results obtained and to improve external
validity.
Nonconcurrent methodological triangulation occurs when a statistical or other machine learning
method is used in combination with an ANN, but the other method is applied to data prior to the
ANN to refine the input vector and gain confidence in the reliability of the independent variables
or alternately after the ANN has produced its results to improve the overall performance and/or
interpretation of those results. The triangulated ANN will be applied to the case of estimating an
individual’s computer self-efficacy (CSE) without relying on self-evaluation, since self-
evaluation may be subject to numerous biases

Q10.ANNs have been shown to be able to accurately approximate


almost any type of model for both classification and regression,
including both linear and nonlinear models. Explain
Ans-An Artificial Neural Network (ANN), usually called neural network (NN), is a
mathematical model or computational model that is inspired by the structure and/or functional
aspects of biological neural networks. A neural network consists of an interconnected group of
artificial neurons, and it processes information using a connectionist approach to computation. In
most cases an ANN is an adaptive system that changes its structure based on external or internal
information that flows through the network during the learning phase. Modern neural networks
are non-linear statistical data modeling tools. They are usually used to model complex
relationships between inputs and outputs or to find patterns in data.
In some of these systems, neural networks or parts of neural networks (such as artificial neurons)
are used as components in larger systems that combine both adaptive and non-adaptive elements.
While the more general approach of such adaptive systems is more suitable for real-world
problem solving, it has far less to do with the traditional artificial intelligence connectionist
models. What they do have in common, however, is the principle of non-linear, distributed,
parallel and local processing and adaptation.

You might also like