Peeking Inside DNNs With Information Theory
Deep learning model performance has taken huge strides, allowing researchers to tackle tasks which were simply not possible for machines less than a decade ago.
Tag: Deep Neural Networks
BatchNorm + Dropout = DNN Success!
A group of researchers from Tencent Technology, the Chinese University of Hong Kong, and Nankai University recently combined two commonly used techniques — Batch Normalization (BatchNorm) and Dropout — into an Independent Component (IC) layer inserted before each weight layer to make inputs more independent*.
Google AI ‘Translatotron’ Can Make Anyone a Real-Time Polyglot
Google AI yesterday released its latest research result in speech-to-speech translation, the futuristic-sounding “Translatotron.”
TableBank: Benchmark for Image-based Table Detection and Recognition
A collaboration between researchers from China’s Beihang University and Microsoft Research Asia has produced TableBank, a new image-based dataset for table detection and recognition built with novel weak supervision from Word and Latex documents on the Internet.
Microsoft’s New MT-DNN Outperforms Google BERT
Microsoft researchers have released technical details of an AI system that combines both approaches. The new Multi-Task Deep Neural Network (MT-DNN) is a natural language processing (NLP) model that outperforms Google BERT in nine of eleven benchmark NLP tasks.
Global Minima Solution for Neural Networks?
New research from Carnegie Mellon University, Peking University and the Massachusetts Institute of Technology shows that global minima of deep neural networks can been achieved via gradient descent under certain conditions. The paper Gradient Descent Finds Global Minima of Deep Neural Networks was published November 12 on arXiv.
Customer Success with Machine Learning (Google Cloud NEXT’17)
The reason why TensorFlow is so widely used is due to its automatic derivation of functions and distributed computing capability. This increase in interest can also be found in commercial applications.
How a Deep Neural Network Trained with End-to-End Learning Steers a Car
A practical way to make an autonomous vehicle is not by programming a car to drive in any environment, but by showing the car how to drive and make the car learn by itself. NVIDIA created a system of this kind, named PilotNet.
Universal Schema: Bridging the Gap Between Strings and Things
Andrew McCallum explained how the use of Universal schema can improve knowledge representation and reasoning from natural language.
The Future of Computing: Neuromorphic
Neuromorphic computing, a promising direction for future of computing that emulates human brain to maintain energy efficiency and fast computing speed.
London: Deep Learning in Healthcare
At RE•WORK Summits, speakers are invited to present advances from the world’s leading innovators, showcase the opportunities in emerging health care industry
Statistical Spoken Dialogue Systems and the Challenges for Machine Learning
This talk describes the dialog system architecture and explains the three main steps of the architecture: understanding, generation, and dialog manager and their challenges for machine learning.
David Silver, Google DeepMind: Deep Reinforcement Learning
There are three ways to combine DL and RL, based on three different principles: value-based, policy-based, and model-based approaches with planning.
Andrej Karpathy CS294 Lecture Review: Visualizing and Understanding Deep Neural Networks
This review will go over some of the current methods that are used to visualize and understand deep neural networks.