GitHub

TensorFlow implementation of the Value Iteration Networks #NIPS2016 paper  #DeepLearning

  • TensorFlow implementation of the Value Iteration Networks (NIPS ’16) paper
  • The GridWorld dataset used is from the author’s repository.
  • The repository contains an implementation of Value Iteration Networks in TensorFlow which won the Best Paper Award at NIPS 2016.
  • The code is based on the original Theano implementation by the authors.
  • The code currently runs the 8×8 GridWorld model by default.

tensorflow-value-iteration-networks – TensorFlow implementation of the Value Iteration Networks (NIPS ’16) paper

@kdnuggets: TensorFlow implementation of the Value Iteration Networks #NIPS2016 paper #DeepLearning

This repository contains an implementation of Value Iteration Networks in TensorFlow which won the Best Paper Award at NIPS 2016. This code is based on the original Theano implementation by the authors.

. The code currently runs the 8×8 GridWorld model by default.

The 8×8 GridWorld model converges in under 30 epochs with about ~98.5% accuracy. The paper lists that it should be around 99.6% and I was able to reproduce this with the Theano code. The TensorFlow model is not perfect as NaNs result when training with the same parameters as the Theano implementation on the 16×16 and 28×28 domain.

GitHub