Deep Learning Weekly | Issue #96: AI and the News Challenge, InteriorNet DataSet, Reinforcement Learning Tutorial, Neural Processes in Pytorch

Bringing you everything new and exciting in the world of
 deep learning from academia to the grubby depth
 of industry every week right to your inbox. Free.

 
Howdy folks and welcome to another exciting week in deep learning! This week we have an open challeng
 
September 21 · Issue #96 · View online
Deep Learning Weekly
Howdy folks and welcome to another exciting week in deep learning!
This week we have an open challenge for you on AI in the news, Denny Britz compiled an extensive companion to your reinforcement learning studies, a blog post by Kaggle lays out useful tips for when you’re unable to reproduce a result from a paper and a fruitful collaboration between a Chinese startup and the Imperial College London produces InteriorNet a mega-scale multi-sensor photo-realistic indoor scenes dataset.
As always happy reading and hacking!

Industry
Machine Learning Confronts the Elephant in the Room
AI and the News: An Open Challenge
Nvidia Launches the Tesla T4, its Fastest Data Center Inferencing Platform Yet
Microsoft Acquires Lobe to Help Bring AI Development Capability to Everyone
Sponsored Link
Learning
Implementation of Reinforcement Learning Algorithms. Python, OpenAI Gym, Tensorflow
Deep Learning models: Object detection with Go using TensorFlow
Analysis of Deep Neural Networks for Pixel Processing
Help! I Can’t Reproduce a Machine Learning Project!
Neural Processes in PyTorch
Datasets
InteriorNet: Mega-scale Multi-sensor Photo-realistic Indoor Scenes Dataset
Libraries & Code
ProSR - A Fully Progressive Approach to Single-Image Super-Resolution
GitHub - SimonKohl/probabilistic_unet: A U-Net combined with a variational auto-encoder that is able to learn conditional distributions over semantic segmentations.
Papers & Publications
Hamiltonian Descent Methods
Phrase-Based & Neural Unsupervised Machine Translation
CGIntrinsics: Better Intrinsic Image Decomposition through Physically-Based Rendering