In this article, you will learn 84 Advanced Deep learning Interview questions and answers for freshers, experienced professionals, AI Engineers and data scientists. If you are not still yet completed machine learning and data science. Here is the list of machine learning interview questions, data science interview questions, python interview questions and sql interview questions.

What is deep learning ?

What is difference between machine learning and deep learning ?

**What are the prerequisites for starting out in Deep Learning?**

Machine learning, Mathematics, Statistics, python programming

**When we will use deep learning ?**

Deeplearning models are used for complex problems and data is huge.

What are advantages of deep learning ?

**What are applications of deep learning ?**

Computer vision, natural language processing and pattern recognition, Image recognition and processing, Machine translation, Sentiment analysis, Question Answering system, Object Classification and Detection, Automatic Handwriting Generation, Automatic Text Generation.

**What are disadvantages of deep learning ? **

These model takes longer time to execute the models. Sometimes it takes days to execute a single model depends on complexity.

These deep learning models will fail on small data sets

**What are deep learning frameworks or tools ?**

Tensorflow, Keras, Pytorch, Theano & Ecosystem,Caffe2, Chainer, CNTK, DSSTNE, DyNet Gensim, Gluon, ,Mxnet, Paddle, BigDL

**What is supervised learning ?**

When we training model with labeled data. Then it’s called supervised learning.

**What is unsupervised learning ?**

When we training model with unlabeled data. Then it’s called unsupervised learning.

What is reinforcement learning ?

What is difference between deep learning and reinforcement learning ?

What is feature engineering ?

Do we require feature extraction in deep learning ?

### Neural network interview questions

What is Neural network ?

**How many types of Neural networks ?**

- Feedforward Neural Network
- Radial basis function Neural Network
- Kohonen Self Organizing Neural Network
- Recurrent Neural Network ( Backforward Neural network)
- Convolutional Neural Network
- Modular Neural Network
- Deep belief networks (Boltzmann Machine)
- Auto Encoders
- Recursive Neural Network

**How many types of Recurrent neural Networks are there in deep learning?**

- CTRNN – continuous-time recurrent neural network
- LSTM – Long short term memory
- GRU – Gated recurrent unit
- Bi directional – recurrent neural network
- Multiple timescales recurrent neural network
- Hierarchical – recurrent neural network

**What are supervised learning algorithms in Deep learning?**

- Artificial neural network
- Convolution neural network
- Recurrent neural network

**What are unsupervised learning algorithms in deep learning ?**

- Self Organizing Maps
- Deep belief networks (Boltzmann Machine)
- Auto Encoders

**What is neural network structure or How many layers in neural network ?**

- Input layer
- Hidden layer
- Output layer

What are weights in neural network ?

What is dense layer ?

What is activation function ?

**How many types of activation functions are there ?**

- Binary Step
- Sigmoid
- Tanh
- ReLU
- Leaky ReLU
- Randomized Leaky Relu
- Softmax

What is binary step function ? formula, When we will use this ?

What is Sigmoid function? formula, When we will use this ?

What is Tanh function? formula, When we will use this ?

What is ReLU function? formula, When we will use this ?

What is Leaky ReLU Function ? formula, When we will use this ?

What is Randomized Leaky Relu function ? formula, When we will use this ?

What is Softmax function ? formula, When we will use this ?

**What is mostly used activation function ?**

Relu function is mostly used activation function. It will help us to solve vanishing gradient problem

**Can i use Relu activation function in output layer ?**

No – it has to use in hidden layers

**In which layer softmax action function will be used ?**

Output layer

What is Dropout function?

What is Cost function ?

What is error rate ? How can you decrease it ?

What is epcoh ?

What is batch size ?

What is Regularization ?

What is learning rate ? What is use of it ?

What is overfitting ? How can you recover from it in deep learning ?

What is underfitting ? How can you recover from it in deep learning ?

How many types of Gradient problems are there ?

- Exploding gradient problem
- Vanishing gradient problem

What is Exploding gradient problem, when it will occur ? how to solve it ?

What is Vanishing gradient problem, when it will occur ? how to solve it ?

What is feed forward network ?

What is back forward network ?

Why are GPUs necessary for building Deep Learning models?

What is the future of Deep Learning?

What Convolutional neural network ?

What are layer in CNN ?

What is pooling layer ? When we can use it ?

**Types of pooling in CNN ?**

- Mean pooling
- Max pooling
- Sum pooling

What is advantages and applications of CNN?

What is disadvantages of CNN?

What is text classification ?

What is sequence model ?

What is RNN ?

What is advantages of RNN ?

What is disadvantages of RNN?

What is Long term short term memory network ?

Why we need to use LTSM over RNN in sequence modeling ?

What is encoder ?

What is decoder ?

What is Radial basis function Neural Network ?

What is Kohonen Self Organizing Neural Network ?

What is Modular Neural Network ? When we will use this ?

What is Deep belief networks ? When we will use this ?

What is Boltzmann Machine network ? When we will use this ?

What are Auto Encoders ? When we will use this ?

What is Recursive Neural Network ? When we will use this ?

### Tensorflow interview questions and answers

What is Tensorflow ?

Advantages of Tensorflow ?

### Keras interview questions and answers ?

What is Keras ?

Why we should use keras ?

**How many types of optimizers are there in keras ? **

- SGD – Stochastic gradient descent optimizer
- RMSprop- RMSProp optimizer
- Adagrad
- Adadelta
- Adam
- Adamax
- Nadam- Nesterov Adam optimizer

**What is transfer learning ?**

a deep learning model trained on a specific task can be reused for different problem in the same domain even if the amount of data is not that huge.

**How can you do object recognition ?**

**What are pre-trained models available in keras ?**

- Xception
- VGG16
- VGG19
- ResNet50
- InceptionV3
- InceptionResNetV2
- MobileNet
- DenseNet
- NASNet
- MobileNetV2

**What are application of deep learning in Natural language processing ?**

Machine translation, Sentiment analysis, Question Answering system, Voice research and recognition.

### Computer vision interview questions and answers

What is computer vision ?

What are application of computer vision ?

How can you do face detection in computer vision ?

How can you do object detection in computer vision ?

How can you input images in computer vision ?

How can you videos images in computer vision ?

Which is best algorithm for face recognition ?

How can you decide which algorithm is suitable for your requirement ?

How can you improve your model accuracy ?

How to deploy deep learning model ?

Explain about one algorithm, which you have used in your projects from start to end ?