Day 1: What is Deep Learning?

What is Deep Learning?

Deep Learning is a subset of Machine Learning that works by mimicking the working of a human brain.

Wikipedia in its definition for deep learning quotes

Deep learning (also known as deep structured learning) is part of a broader family of machine learning methods based on artificial neural networks with representation learning. Learning can be supervised, semi-supervised or unsupervised.

Note that machine learning is in turn a subset of Artificial Intelligence.

What are the different types of Deep Learning Architectures?

A deep learning architecture can be of various types. There is a range of neural networks which are used for different problem statements. Some of them include:

1. Deep Neural Networks
2. Graph Neural Networks
3. Recurrent Neural Networks
4. Convolutional Neural Networks

and more

What is the difference between Machine Learning and Deep Learning?

• Machine Learning uses algorithms to analyze data, understand the data, and make informed decisions based on what it has analyzed.
• Deep Learning is a subset of Machine Learning.
• Deep Learning uses artificial neural networks that learn through data and make intelligent decisions.
• Deep Learning requires very large datasets as compared to Machine Learning although it can be argued by the use of transfer learning.
• Machine Learning models can be trained on the CPU, but Deep Learning requires GPU for proper training.

Use cases of deep learning?

Different architectures are associated with multiple applications. For eg. Convolution Neural Networks are widely used for Computer Vision (image and video data) based products similarly Recurrent Neural Networks are associated with Natural Language Processing (text data). In this manner, one can easily comprehend that deep learning can be used in a variety of industries on mostly all types of structure data. Use cases of deep learning include

1. Computer Vision
   1. Image classification
   2. Object detection
2. Natural Language Processing
   1. Text classification
   2. Neural Search
   3. Entity Recognition
   4. Machine Translation
3. Recommendation Engines
4. Audio processing

Components of Artificial Neural Networks

Neural networks

As the name suggests Neural Networks can be defined as a set of algorithms that are specifically designed to act like a human brain. In a technological manner, it can be referred to as a computational model that has a network architecture.

We will cover them in detail in upcoming blogs. To get a general idea here some of its components.

• Neurons
• Layers of Neural Network
  1. Input Layer
  2. Hidden Layers
  3. Output Layer

• Weights & Biases
• Loss functions
• Activations functions
• Optmisation / Gradient Descent
• Parameter Tuning

References

• https://en.wikipedia.org/wiki/Deep_learning
• https://en.wikipedia.org/wiki/Neural_network
• https://en.wikipedia.org/wiki/Artificial_neural_network