Seven important approaches in deep learning

Seven important types/concepts/approaches in deep learning:

  1. Feed Forward Neural Networks (FFNNs) – classification and regression based on features. See Part 1 of this tutorial for an example. [Supervised]
  2. Convolutional Neural Networks (CNNs) – image classification, object detection, video action recognition, etc. See Part 2 of this tutorial for an example. [Supervised]
  3. Recurrent Neural Networks (RNNs) – language modeling, speech recognition/generation, etc. See this TF tutorial on text generation for an example. [Supervised]
  4. Encoder Decoder Architectures – semantic segmentation, machine translation, etc. See our tutorial on semantic segmentation for an example. [Supervised]
  5. Autoencoder – unsupervised embeddings, denoising, etc. [Unsupervised]
  6. Generative Adversarial Networks (GANs) – unsupervised generation of realistic images, etc. See this TF tutorial on DCGANs for an example. [Unsupervised]
  7. Deep Reinforcement Learning – game playing, robotics in simulation, self-play, neural arhitecture search, etc. We’ll be releasing notebooks on this soon and will link them here. [Reinforcement]

Source: tutorial_deep_learning_basics.ipynb – Colaboratory

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.