Versions Compared

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

Higher order gradient calculation is required for many applications such as adaptive learning rate optimization [1], WWGAN-GAN GP network [2], network neural architecture search [3], etc. Implementing higher order gradient can help unlock these applications and improve the usability and popularity of Apache MXNet framework.

...

[1] Forward and Reverse Gradient-Based Hyperparameter Optimization

[2] Improved Training of Wasserstein GANGANs

[3] Neural Architecture Search with Reinforcement Learning

...