Keyword spotting (KWS) constitutes a major component of human-technology interfaces. Maximizing the detection accuracy at a low false alarm (FA) rate, while minimizing the footprint size, latency and complexity are the goals for KWS. Towards achieving them, we study Convolutional Recurrent Neural Networks (CRNNs). Inspired by large-scale state-of-the-art speech recognition systems, we combine the strengths of convolutional layers and recurrent layers to exploit local structure and long-range context. We analyze the effect of architecture parameters, and propose training strategies to improve performance. With only ~230k parameters, our CRNN model yields acceptably low latency, and achieves 97.71% accuracy at 0.5 FA/hour for 5 dB signal-to-noise ratio.

Go, an ancient Chinese game that's notorious for its strategic difficulty, is a critical milestone for artificial intelligence. Computers have been beating people at chess since 1997 by efficiently searching through possible moves. But Go is a whole different beast; to win, a computer has to think more like a person using their intuition. So when the AlphaGo program designed by DeepMind beat Lee Sedol, one of the world's best Go players, in 2016, it was astonishing. Even more astonishing, it played in a way that was surprising and original1 to experienced Go players.

New tutorial on unsupervised pre-training with stacked denoising auto-encoders. An IJulia Notebook demo of using pre-trained CNN on imagenet to do image classification.

A British hospital trust misused patient data when it shared information with Google for work on a smartphone app, a British data protection watchdog said on Monday. The Royal Free NHS Trust did not comply with the Data Protection Act when it passed on personal information of around 1.6 million patients to Google's DeepMind. DeepMind, which is owned by Alphabet, the parent company of Google, is the company's artificial intelligence and machine learning branch. The Information Commissioner's Office (ICO) said it found'a number of shortcomings' in the way the data was handled, including that patients were not adequately informed their data would be used as part of the test. A British hospital trust misused patient data when it shared information with Google's DeepMind for work on a smartphone app, a British data protection watchdog has found.

A few months ago, we showed how effectively an LSTM network can perform text transliteration. For humans, transliteration is a relatively easy and interpretable task, so it's a good task for interpreting what the network is doing, and whether it is similar to how humans approach the same task. In this post we'll try to understand: What do individual neurons of the network actually learn? How are they used to make decisions? About half of the billions of internet users speak languages written in non-Latin alphabets, like Russian, Arabic, Chinese, Greek and Armenian. Very often, they haphazardly use the Latin alphabet to write those languages.

A British regulatory organization has found that the National Health Service violated data privacy laws when it shared patient records with Google's DeepMind artificial intelligence startup. In a statement announcing its findings, the Information Commissioner's Office said the Royal Free NHS Foundation Trust did not comply with the Data Protection Act when it provided partial records for more than 1.6 million patients to DeepMind. The data was originally provided to help bolster DeepMind's Streams app and improve detection of acute kidney injury and other medical problems. However, the Information Commissioner's Office found that the Foundation Trust should have taken additional measures to inform patients about the data use. In its release, commissioner Elizabeth Denham said Streams and DeepMind's work had clear benefits, but the Trust should have been clearer about the amount of data it needed and the reasons it wanted patient data.

This package also comes with popular loss functions (e.g.

"The cutting-edge of autonomous driving has shifted squarely to deep learning. Even traditional autonomous driving teams have'sprinkled on' some deep learning, but Drive.ai is at the forefront of leveraging deep learning to build a truly modern autonomous driving software stack. "The leap from transactional automation to cognitive automation is imminent and it will forever transform the way we work," says Frederic Laluyaux, President and CEO of Aera. "At Aera, we deliver the technology that enables the Self-Driving Enterprise: a cognitive operating system that connects you with your business and autonomously orchestrates your operations." Said Luis Dussan, CEO of AEye. "The biggest bottleneck to the rollout of robotic vision solutions has been the industry's inability to deliver a world-class perception layer.

An NHS Trust broke the law by sharing sensitive patient records with Google's DeepMind division, the UK's data watchdog has ruled. The long-awaited decision falls in line with the conclusion drawn by Dame Fiona Caldicott, the UK's National Data Guardian in May. The pair's agreement "failed to comply" with the Data Protection Act 1998, according to the Information Commissioner's Office (ICO), because patients weren't informed that their information was being used. The ICO also took issue with the size of the dataset -- 1.6 million partial patient records -- leveraged by DeepMind to test Streams, an app for detecting acute kidney injury. In April 2016, New Scientist revealed that DeepMind and Royal Free London NHS Trust were working together on a medical project. As the ICO notes in its letter to the Trust, their agreement was actually formalised in September 2015, with Royal Free serving as the data controller (owner) and DeepMind as the data processor (partner).

Snap Inc. is a camera company. We believe that reinventing the camera represents our greatest opportunity to improve the way people live and communicate. Our products empower people to express themselves, live in the moment, learn about the world, and have fun together.