Skip to content

Latest commit

 

History

History
29 lines (20 loc) · 2.43 KB

Deep Learning.md

File metadata and controls

29 lines (20 loc) · 2.43 KB

  • Similar to a traditional neural network, but it has many more hidden layers.

  • Deep learning has emerged now because of the following reasons:

  • Emergence of big data, which requires data processing scaling.

  • Improvement in processing power and the usage of GPUs to train neural networks.

  • Advancement in algorithms like the rectified linear unit (ReLU).

Deep learning is a machine learning technique that uses neural networks to learn. Although deep learning is similar to a traditional neural network, it has many more hidden layers. The more complex the problem, the more hidden layers there are in the model.

Deep learning has emerged now because of the following reasons:

  • The continuous increase in big data requires data processing scaling to analyze and use this data correctly.
  • Improvement in processing power and the usage of GPUs to train neural networks.
  • Advancement in algorithms like the rectified linear unit (ReLU) instead of the Sigmoid algorithm helps make gradient descent converge faster.

Applications :

  • Multilayer perceptron (MLP): Classification and regression, for example, a house price prediction.
  • Convolutional neural network (CNN): For image processing like facial recognition.
  • Recurrent neural network (RNN): For one-dimensional sequence input data. Like audio and languages.
  • Hybrid neural network: Covering more complex neural networks, for example, autonomous cars.

DL Algorithms

  • Multilayer perceptron (MLP): A class of feed-forward artificial neural networks (ANNs). It is useful in classification problems where inputs are assigned a class. It also works in regression  problems for a real-valued quantity like a house price prediction.
  • Convolutional neural network (CNN): Takes an input as an image. It is useful for image  recognition problems like facial recognition.
  • Recurrent neural network (RNN): Has a temporal nature where the input may be a function in time, such as audio files. It is also used for one-dimensional sequence data. It is suitable for inputs like audio and languages. It can be used in applications like speech recognition and machine translation.
  • Hybrid neural network: Covers more complex neural networks, for example, autonomous cars  that require processing images and work by using radar.