Scribd
Upload
35 views3 pages

Tutorial Instruction Answer The Following Questions

Neural networks can be applied in healthcare for tasks like disease diagnosis, predicting patient outcomes, and processing medical images. They have also been used for applications like face recognition on photos and speech recognition in devices. Neural networks are well-suited for complex, non-linear problems because they can learn relationships in data and make inferences. Some key advantages include storing information across the entire network, working with incomplete knowledge, and performing parallel processing.

Uploaded by

Katoon Ghode
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
35 views3 pages

Tutorial Instruction Answer The Following Questions

Neural networks can be applied in healthcare for tasks like disease diagnosis, predicting patient outcomes, and processing medical images. They have also been used for applications like face recognition on photos and speech recognition in devices. Neural networks are well-suited for complex, non-linear problems because they can learn relationships in data and make inferences. Some key advantages include storing information across the entire network, working with incomplete knowledge, and performing parallel processing.

Uploaded by

Katoon Ghode
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 3

Tutorial Instruction

Answer the following questions:


1. Name some applications of Neural Network (DL in particular) in Healthcare
industry.
(Answer)
Some applications of Neural Network in Healthcare industry:
Applications in basic sciences:
- Identify pathogens in microbiology, and design and discover new drugs in
pharmacology
- Analysis the sequential data in biological structures
- Predict the secondary and tertiary structures of proteins, DNA, and other
biologically interesting sequences
Applications in clinical medicine:
- Make prediction of diagnoses (myocardial infarction, giant cell arteritis, several
types of cancer,…)
- Predict the outcomes, such as length of stay, charges, prognoses, and rate of
complications (valve-related complications in heart disease, length of stay in the
intensive care unit, admission to the psychiatry ward,…)
Applications in image processing:
- Process images in radiology, nuclear medicine, and in other medical areas in
which images are fundamental tools, such as dermatology, pathology, and
endoscopy
2. Name some applications of DL (if) you have meet in real life.
(Answer)
- Image recognition in assigning a name to a photographed face on Google
photos, or “tagging” on Facebook.
- Speech recognition in devices such as Alexa.

3. You know that with normal computational machine, we can solve almost problems.
However, we still need to develop Neural Network. So why do we use Artificial
Neural Networks? What are its advantages?
(Answer)
Neural networks are ideally suited to help people solve complex problems in real-life
situations. They can learn and model the relationships between inputs and outputs that
are nonlinear and complex; make generalizations and inferences; reveal hidden
relationships, patterns and predictions; and model highly volatile data (such as
financial time series data) and variances needed to predict rare events (such as fraud
detection). As a result, neural networks can improve decision processes in areas.
There are various advantages of neural networks:
- Store informtion on the entire network
- The ability to work with insufficient knowledge
- Good falt tolerance
- Distributed memory
- Gradual corruption
- Ability to train machine
- The ability of parallel processing

4. Can you simply explain what is Perceptron?


(Answer)
A perception is a simple model of a biological neuron in an artificial neural network.
Perception is also the name of an early algorithm for supervised learning of inary
classifiers. The perceptron algorithm classifies patterns and groups by finding the
linear separation between different objects and patterns that are received through
numeric or visual input.

5. What is activation function in Neural Network? Why do we need an activation


function in NN?
(Answer)
Activation function decides, whether a neuron should be activated or not by
calculating weighted sum and further adding bias with it. The purpose of the activation
function is to introduce non-linearity into the output of a neuron.
A neural network without an activation function is essentially just a linear regression
model. The activation function does the non-linear transformation to the input making
it capable to learn and perform more complex tasks. This observation results again in a
linear function even after applying a hidden layer, hence we can conclude that, doesn’t
matter how many hidden layer we attach in neural net, all layers will behave same way
because the composition of two linear function is a linear function itself. Neuron can
not learn with just a linear function attached to it. A non-linear activation function will
let it learn as per the difference w.r.t error. Hence we need activation function.

6. Name some popular activation function.


(Answer)
- Binary step function
- Sigmoid function
- ReLU function
7. What is hidden layers in NN? Why is it called “Hidden”?
(Answer)
A hidden layer in an artificial neural network is a layer in between input layers and
output layers, where artificial neurons take in a set of weighted inputs and produce an
output through an activation function. It is a typical part of nearly any neural network
in which engineers simulate the types of activity that go on in the human brain.
It’s called “hidden” because it is not directly observable from the systems inouts and
outputs.
8. How many hidden layers in Neural Network?
(Answer)
There are many hidden layers in Neural Network. Determining the number of hidden
layers to use with the neural network is not easy. Two or fewer layers will often
suffice with simple data sets.
- No layer: Only capable of representing linear separable functions or decisions.
- 1 layer: Can approximate any function that contains a continuous mapping from
one finite space to another.
- 2 layers: Can represent an arbitrary decision boundary to arbitrary accuracy
with rational activation functions and can approximate any smooth mapping to
any accuracy.
- More than 2 layers: Additional layers can learn complex representations (sort of
automatic feature engineering) for layer layers.

You might also like