In this article, you will learn 84 Advanced Deep learning Interview questions and answers for freshers, experienced professionals, AI Engineers and data scientists.

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 ?