Top 45 Deep Learning Interview Questions and Answers

What is Deep Learning?


Deep Learning involves networks which are capable of learning from data and functions similar to the human brain.

Deep Learning Interview Questions

Why we need Deep Learning?

1) To Processes massive amount of data
Deep Learning can process an enormous amount of both Structured and Unstructured data.

2) To Performs Complex Operations
Deep Learning algorithms are capable enough to perform complex operations when compared to the Machine Learning algorithms.

3) To Achieves Best Performance

As the amount of data increases, the performance of Machine Learning algorithms decreases.
On the other hand, Deep Learning maintains the performance of the model.

4) To Feature Extraction

Deep Learning vs. Machine Learning

Machine Learning

If done through Machine Learning, we need to specify the features based on which the two can be differentiated like size and stem, in this case.

Deep Learning

In Deep Learning, the features are picked by the Neural Network without any human intervention. But, that kind of independence can be achieved by a higher volume of data in training the machine.

Neural Networks

A Node is also called a Neuron or Perceptron.

Structure of an Artificial Neural Network

Structure of an Artificial Neural Network

Neural Networks


Hidden Layer

  • A Neural Network consists of several hidden layers, each consisting of a collection of neurons.
  • The hidden layer is an intermediate layer found between the input layer and the output layer.
  • This layer is responsible forextracting the features required from the input.
  • There is no exact formula for calculating the number of the hidden layers as well as the number of neurons in each hidden layer.

  • Output Layer
  • The output layer of the neural network collects and transmits information in the desired format.

Neural Network Structure


Summation and Activation Functions


Summation Function

Various inputs are multiplied with their respective connection weights and summed up together with the bias value.
\sum_{j=1}^{N}X_jW_j+b_k∑
​j=1
​N
​​ X
​j
​​ W
​j
​​ +b
​k
​​

where:

X_jX
​j
​​ represents the inputs

W_jW
​j
​​ represents the weights

Activation Function

  • Activation Function aids in deriving the output.
  • It is also known as the Transfer Function.
  • It maps the resulting values between 0 to 1 or -1 to 1.

There are two types of Activation function, namely:

Linear Activation Function
Non-Linear Activation Function
Non-Linear Activation Function is the most commonly used Activation function in Neural Networks.

Learning Process of a Neural Network
The learning process of a Neural Network includes updating the network architecture and connecting the weights for the network to perform efficiently.

Designing a Learning Process involves the following:

Learning Paradigm
Having a model from the environment in which Neural Network works.

Learning Rules
Figuring out the rules that aid in updating the weights.

Learning Algorithms
Identifying the procedure to update the weights according to the learning rules.

Learning Paradigm
The following are the various Learning Paradigms in Neural Networks:

  • Supervised
  • Unsupervised
  • Reinforcement

Learning Rules
The four basic types of Learning Rules in Neural Network are:

  • Error Correction Rules
  • Hebbian
  • Boltzmann
  • Competitive Learning

Learning Algorithms
Following are a few Deep Learning algorithms:

  • Adaptive Resonance Theory
  • Kochen Self Organization Map
  • ADALINE
  • Perceptron
  • Backpropagation

Learning Principles
A Neural Network works based on two principles, namely:

  • Forward Propagation
  • Backward Propagation

Neural Network Architecture
The architecture of a Neural Network can be broadly classified into two, namely:

Neural Network architecture
  • Feed Forward Artificial Neural Network
  • Recurrent Neural Network

The information must flow from input to output only in one direction.
No Feedback loops must be present.
A few Feed Forward Artificial Neural Networks are:

  • Single-layer Feed Forward Network
  • Multi-layer Feed Forward Network

Neural Networks are employed in various fields. Following are a few types of Neural Networks that we will explore in this section:

  • Radial Basis Function
  • Long/Short Term Memory
  • Gated Recurrent Unit
  • Autoencoder
  • Convolutional Neural Networks


Machine Learning algorithms extract patterns from labeled sample data, while Deep Learning algorithms take large volumes of data as input, analyze them to extract the features on its own.

Neural Networks are employed in various fields. Following are a few types of Neural Networks that we will explore in this section:

  • Radial Basis Function
  • Long/Short Term Memory
  • Gated Recurrent Unit
  • Autoencoder
  • Convolutional Neural Networks

Radial Basis Function

Long/Short Term Memory
Long/Short Term Memory (LSTM) networks are used to classify, process, and make predictions based on time series dat(i)
LSTM networks can predict the action in a specific video frame by keeping in mind the action that occurred in the earlier frames.
The applications of LSTM include writing, speech recognition, and so on.

Gated Recurrent Unit
Gated Recurrent Unit (GRU) is LSTM with a forget gate.
It is used in sound, speech synthesis, and so on.

Image Classification
Image classification is one of the common applications of deep learning.

A convolutional neural network can be used to recognize images and label them automatically.

Deepnets (an optimized version of Deep Neural Networks) can be trained to recognize different objects within the same image.

Sentiment Analysis
Using sentiment analysis, the underlying intent of the text can be extracte(iv)

With social media channels, it is possible to automate and measure the feelings of the public on a given news story, topic, brand, or product.

Positive sentiment can be identified, thereby allowing the marketing of a product, or understanding which elements of a business strategy are working.

Medical Applications
Deepnets can be trained to detect cancerous cells, benign and malignant tumors from MRI and CT scans.

They are also applied in drug discovery by training nets with molecular structure and chemical compositions.

Deepnet Platform


A deepnet platform is a service that allows you to incorporate deepnet in your applications without building one from scratch.
This platform provides a set of tools and interfaces to create a custom deepnet.


Deepnet platforms are of two types, namely:

Software platform: This platform is available as downloadable packages that need to be deployed on your hardware.
Full platform: It is available as an online interactive UI to build and deploy models without any coding experience.

Deepnet Platforms – Tools


The following are the tools offered by the deepnet platforms:

  • Deepnet capability
  • Data Munging
  • UI/Model Management
  • Infrastructure


H_2O.ai

​​ O.ai is an open-source machine learning platform.
Along with many machine learning algorithms, this platform currently provides a deepnet capability known as Multilayer Perceptron (MLP).


H_2O.aiH Features

H_2O.aiH
​2
​​ O.ai supports MLP for a deep learning model, which uses the L-BFGS algorithm for backpropagation.
H_2O.aiH
​2
​​ O.ai offers built-in integration tools for platforms like HDFS, Amazon S3, SQL, and NoSQL.
Its intuitive UI can be accessed by programming environments like R, Python, and JSON.
You can model and analyze data with Tableau, Microsoft Excel, and RStudio.
Since the package needs to be downloaded and deployed in the hardware, H_2O.aiH
​2
​​ O.ai comes with an in-memory map-reduce, and columnar compression to accelerate the training of models.

Turi – A Graph-based Tool


Turi is a software platform that provides two deepnet capabilities (convolutional network and MLP) along with machine learning and graph algorithms.

Features

Turi provides built-in support for integration of Amazon S3, SQL DB, HDFS, Spark RDD, and Pandas data frames.
It offers a UI interface for model management and also includes a visualization tool called GraphLab Canvas for visualizing the model results.
To work with large data, Turi comes with built-in storage support like SFrame, SArray, and SGraph.
Unlike H_2O.aiH
​2
​​ O.ai, Turi supports the use of GPUs.

Deep Learning Questions Answers

Deep Learning Interview Questions

Below we have tried to collect some Deep Learning Multiple Choice Questions based on the above article. These Advanced Deep Learning Questions includes Artificial Intelligence and Neural Network Interview Questions as well. Please Share and Like if you like these Deep Learning MCQ Questions , we will try to add more for the same.

Below are the different Deep Leaning Questions and answer are followed by the questions

(1)Recurrent Neural Networks are best suited for Text Processing.
i) True
ii) False
Answer:-True

(2)Which of the following is/are Common uses of RNNs?

i) BusinessesHelp securities traders to generate analytic reports
ii) Detect fraudulent credit-card transaction
iii) Provide a caption for images
iv) All of the above
Answer:- All of the above

(3)Prediction Accuracy of a Neural Network depends on ___ and __.
i) Input and Output
ii) Weight and Bias
iii) Linear and Logistic Function
iv) Activation and Threshold
Answer:-Weight and Bias

(4)Which of the following is a subset of machine learning?

i) Numpy
ii) SciPy
iii) Deep Learning
iv) All of the above
Answer:- Deep Learning

(5)GPU stands for __.
i) Graphics Processing Unit
ii) Gradient Processing Unit
iii) General Processing Unit
iv) Good Processing Unit.
Answer:- Graphics Processing Unit

(6)What is the difference between the actual output and generated output known as?
i) Output Modulus
ii) Accuracy
iii) Cost
iv) Output Difference
Answer:-Cost

(7)Gradient at a given layer is the product of all gradients at the previous layers.
i) False
ii) True
Answer:- True

Deep learning questions on convolutional neural networks

(8)_________ is a Neural Nets way of classifying inputs.
i) Learning
ii) Forward Propagation
iii) Activation
iv) Classification
Answer:- Forward Propagation

(9)Name the component of a Neural Network where the true value of the input is not observed.
i) Hidden Layer
ii) Gradient Descent
iii) Activation Function
iv) Output Layer
Answer:- Hidden Layer

(10)____ works best for Image Data.
i) AutoEncoders
ii) Single Layer Perceptrons
iii) Convolution Networks
iv) Random Forest
Answer:- Convolution Networks

(11)Which of the following is well suited for perceptual tasks?

i) Feed-forward neural networks
ii) Recurrent neural networks
iii) Convolutional neural networks
iv) Reinforcement Learning
Answer:- Convolutional neural networks

(12)Neural Networks Algorithms are inspired from the structure and functioning of the Human Biological Neuron.
i) False
ii) True
Answer:- True

(13)In a Neural Network, all the edges and nodes have the same Weight and Bias values.
i) True
ii) False
Answer:- False

(14)Recurrent Networks work best for Speech Recognition.
i) True
ii) False
Answer:-True

(15)___ is a recommended Model for Pattern Recognition in Unlabeled Data.
i) CNN
ii) Shallow Neural Networks
iii) Autoencoders
iv) RNN
Answer:- Autoencoders

(16)Process of improving the accuracy of a Neural Network is called ___.
i) Forward Propagation
ii) Cross Validation
iii) Random Walk
iv) Training
Answer:- Training

(17)What are the two layers of a Restricted Boltzmann Machine called?
i) Input and Output Layers
ii) Recurrent and Convolution Layers
iii) Activation and Threshold Layers
iv) Hidden and Visible Layers
Answer:- Hidden and Visible Layers

(18)A Shallow Neural Network has only one hidden layer between Input and Output layers.
i) False
ii) True
Answer:- True

(19)CNN is mostly used when there is an?

i) structured data
ii) unstructured data
iii) Both A and B
iv) None of the above
Answer:- unstructured data

Advanced Deep learning Interview questions in 2021

20)The rate at which cost changes with respect to weight or bias is called ______.
i) Derivative
ii) Gradient
iii) Rate of Change
iv) Loss

(21)What does LSTM stand for?
i) Long Short Term Memory
ii) Least Squares Term Memory
iii) Least Square Time Mean
iv) Long Short Threshold Memory
Answer:-Long Short Term Memory

(22)All the Visible Layers in a Restricted Boltzmannn Machine are connected to each other.
i) True
ii) False
Answer:- False

(23)All the neurons in a convolution layer have different Weights and Biases.
i) True
ii) False
Answer:- False

(24)Support Vector Machines, Naive Bayes and Logistic Regression are used for solving _______ problems.
i) Clustering
ii) Classification
iii) Regression
iv) Time Series
Answer:- Classification

(25)What is the method to overcome the Decay of Information through time in RNN known as?
i) Back Propagation
ii) Gradient Descent
iii) Activation
iv) Gating
Answer:- Gating

(26)Recurrent Network can input Sequence of Data Points and Produce a Sequence of Output.
i) False
ii) True
Answer:- True

(27)A Deep Belief Network is a stack of Restricted Boltzmann Machines.
i) False
ii) True
Answer:-True

(28)Restricted Boltzmann Machine expects the data to be labeled for Training.
i) False
ii) True
Answer:- False

(29)De-noising and Contractive are examples of ______.
i) Shallow Neural Networks
ii) Autoencoders
iii) Convolution Neural Networks
iv) Recurrent Neural Networks
Answer:-Autoencoders

(30)Which neural network has only one hidden layer between the input and output?
i) Shallow neural network
ii) Deep neural network
iii) Feed-forward neural networks
iv) Recurrent neural networks
Answer:- Shallow neural network

(31)RELU stands for __________________.
i) Rectified Linear Unit
ii) Rectified Lagrangian Unit
iii) Regressive Linear Unit
iv) Regressive Lagrangian Unit
Answer:- Rectified Linear Unit

(32)Why is the Pooling Layer used in a Convolution Neural Network?
i) They are of no use in CNN.
ii) Dimension Reduction
iii) Object Recognition
iv) Image Sensing
Answer:- Dimension Reduction

(33)How many layers Deep learning algorithms are constructed?
i) 2
ii) 3
iii) 4
iv) 5
Answer:- 3

(34)The measure of Difference between two probability distributions is know as ____________.
i) Probability Difference
ii) Cost
iii) KL Divergence
iv) Error
Answer:- KL Divergence

(35)What is the best Neural Network Model for Temporal Data?
i) Recurrent Neural Network
ii) Convolution Neural Networks
iii) Temporal Neural Networks
iv) Multi Layer Perceptrons
Answer:- Recurrent Neural Network

(36)Data Collected from Survey results is an example of _______.
i) Data
ii) Information
iii) Structured Data
iv) Unstructured Data
Answer:- Structured Data

(37)A _____ matches or surpasses the output of an individual neuron to a visual stimuli.
i) Max Pooling
ii) Gradient
iii) Cost
iv) Convolution
Answer:- Convolution

(38)Which of the following is/are Limitations of deep learning?
i) Data labeling
ii) Obtain huge training datasets
iii) Both A and B
iv) None of the above
Answer:- Both A and B

(39)The rate at which cost changes with respect to weight or bias is called ______.
i) Derivative
ii) Gradient
iii) Rate of Change
iv) Loss
Answer:- Gradient

(40)The first layer is called the?
i) inner layer
ii) outer layer
iii) hidden layer
iv) None of the above
Answer:- inner layer

(41)Autoencoders are trained using _________.
i) Feed Forward
ii) Reconstruction
iii) Back Propagation
iv) They do not require Training
Answer:- Back Propagation

(42)Deep learning algorithms are _ more accurate than machine learning algorithm in image classification.
i) 33%
ii) 37%
iii) 40%
iv) 41%
Answer:- 41%

(43)How do RNTS interpret words?
i) One Hot Encoding
ii) Lower Case Versions
iii) Word Frequencies
iv) Vector Representations
Answer:-Vector Representations

(44)RNNs stands for?
i) Receives neural networks
ii) Report neural networks
iii) Recording neural networks
iv) Recurrent neural networks
Answer:- Recurrent neural networks

(45)Autoencoders cannot be used for Dimensionality Reduction.
i) False
ii) True
Answer:-False

Additional Machine Learning Interview Questions and Answers

In __, information must flow from input to output only in one direction.
Choose the correct answer from below list
(i)Recurrent Neural Network
(ii)Feed Forward Network

Answer:-(ii)Feed Forward Network

Every neuron in the Input Layer represents a/an ___ variable that influences the output.

(i)Dependent
(ii)Independent

Answer- (ii)Independent

Radial Basis Function Neural Network uses ___ function as the Activation Function.
Choose the correct answer from below list
(i)Logistic
(ii)Linear

Answer:-(i)Logistic

Deep Learning can process an enormous amount of ___.
Choose the correct answer from below list
(i)Structured Data
(ii)Both the options
(iii)Unstructured Data

Answer:-(ii)Both the options

Comprehensive Neural Netwok Interview Questions and Answers

Autoencoders are trained without supervision.

(i)True
(ii)False

Answer- (i) True

Which of the following is a type of Recurrent Neural Network?

(i)Hopfield Network
(ii)All the options
(iii)Jordan Network
(iv)Elman Network

Answer:-(ii)All the options

The following are tools offered by deepnet platforms, except __.

(i)H2O.ai
(ii)Deepnet Capability
(iii)Data Munging

Answer- (i)H2O.ai

Machine Learning is a subset of _.

(i)Deep Learning
(ii)Artificial Intelligence

Answer- (ii)Artificial Intelligence

Feedback loops are allowed in ___.

(i)Recurrent Neural Network
(ii)Feed Forward Network

Answer- (i)Recurrent Neural Network

__ is a technique which helps machines to mimic human behavior.

(i)Deep Learning
(ii)Machine Learning
(iii)Artificial Intelligence

Answer- (iii)Artificial Intelligence

H2O.ai offers built-in integration tools for which platform?

(i)NoSQL
(ii)All the options
(iii)HDFS
(iv)Amazon S3

Answer: (ii)All the options

__ activation function is the most commonly used activation function in Neural networks.

(i)Linear
(ii)Non-Linear

Answer: -(ii)Non-Linear

In Deep Learning, __.

(i)Features have to be fed to the model

(ii)Features are picked by the network

Answer- : Features are picked by the network

__ tool supports the use of GPU.
Choose the correct answer from below list
(i)H2O.aiH_{2}O.aiH?2??O.ai
(ii)Turi

Answer:-(ii)Turi

__ function is also known as Transfer Function.
Choose the correct answer from below list
(i)Activation
(ii)Summation

Answer:-(i)Activation

If you have any doubt regarding above Deep Learning Topic or Deep Learning Interview Questions, please let us know.

Leave a Comment