PathNet: Evolution Channels Gradient Descent in Super Neural Networks

For artificial general intelligence (AGI) it would be efficient if multiple users trained the same giant neural network, permitting parameter reuse, without catastrophic forgetting. PathNet is a first step in this direction.…Read More

Hybrid computing using a neural network with dynamic external memory : Nature

This article has been recently published by Deep Mind scholars. Here is the abstract. click on the link if you wish to read on:

"Artificial neural networks are remarkably adept at sensory processing,…Read More

Playing Atari with Deep Reinforcement Learning

The paper presents he first deep learning model to successfully learn control policies directly from high-dimensional sensory input using reinforcement learning. It's a must read for AI scientists

NEURAL ARCHITECTURE SEARCH WITH REINFORCEMENT LEARNING

Neural networks are powerful and flexible models that work well for many difficult learning tasks in image, speech and natural language understanding. However, designing these networks is difficult and time intensive.…Read More

A nice overview of all state of art Neural Network Architectures

For a commentary on this paper, please see this link on Medium. 

Deep Mind's solution for continual learning issues in neural networks

From Deep Mind’s blog:
“Deep neural networks are currently the most successful machine learning technique for solving a variety of tasks including language translation, image classification…Read More