|October 18 · Issue #11 · View online |
Hiya fellow deep learning enthusiasts,
We have loads of goodies in the basket this week, you’ll find out how to run trained Keras models in the browser, why you can confidently use stochastic gradient descent with non-convex RNNs, how to build a prognostic models for breast cancer using deep learning and much more.
As always liking, sharing, recommending, commenting and even lamenting is very much appreciated.
| There is a blind spot in AI research |
Interesting Nature article which points to risks of deployed AI systems that are often drowned out by the hysteria surrounding a potentially threatening ‘super-intelligence’.
| Differentiable neural computers | DeepMind |
Differentiable neural computers combine neural networks and memory systems to make learning machines that can store knowledge quickly and reason about it flexibly. These models can learn from examples like neural networks, but they can also store complex data like computers.
| Novel Tensor Mining Tool to Enable Automated Modeling |
Tensors and tensor decompositions, a powerful set of new data mining tools that can be used to model and extract knowledge from multidimensional data, can be automated for more widespread use in Big Data applications.
| How to Run Text Summarization with TensorFlow – Medium |
Tutorial on running the algorithm recently described by Google.
| Gradient Descent Learns Linear Dynamical Systems – Off the convex path |
Interesting and important result; interpreting recurrent neural networks as dynamic systems stochastic gradient descent successfully learns the parameter an unknown linear dynamical system even though the training objective is non-convex.
| GitHub - transcranial/keras-js: Run trained Keras models in the browser, with GPU support |
keras-js - Run trained Keras models in the browser, with GPU support
| GitHub - scienceai/neocortex: Run trained deep neural networks in the browser or node.js |
neocortex - Run trained deep neural networks in the browser or node.js
| Uncertainty in Deep Learning |
A PhD Thesis on how to obtain uncertainty in deep learning.
| Multi-Task Curriculum Transfer Deep Learning of Clothing Attributes |
Recognising detailed clothing characteristics (fine-grained attributes) in unconstrained images of people in-the-wild.
| Deep Learning Assessment of Tumor Proliferation in Breast Cancer Histological Images |
This study presents a data-driven integrative approach to characterize the severity of tumor growth and spread utilizing multiple biologically salient deep learning classifiers to develop a comprehensive prognostic model.