Get ready for neural networks in your pocket

This is a guest post by George Brostoff, CEO of SensibleVision. It is safe to say that AI has experienced a tremendous growth spurt in 2017. While some of the talk seems to border on marketing hype that has gotten way ahead of the actual functionality being delivered, there is no doubt that AI and machine learning capabilities are expanding across a wide range of use cases. From Google Home all the way up to that gaggle of Power 7

NEC develops automatic optimization technology for deep learning

In an effort to facilitate improvements in recognition accuracy, NEC has developed automatic optimization technology for deep learning. In a statement, the company explains that if deep learning systems become excessively familiar with data, they become unable to accurately recognize data that they have not learned. This “overtraining” results in degradation of recognition accuracy when dealing with data that was not used in the learning process. To prevent overtraining, “regularization” technology is commonly used, which regulates the extent of learning

BrainChip unveils new PCIe accelerator card

BrainChip has unveiled a new PCIe accelerator card that serves as a key accessory for neuromorphic computing, which emulates the workings of the human brain, according to a report by TechSource. Neuromorphic hardware implements neural networks designed to process large amounts of data over a short duration. BrainChip’s new neuromorphic accelerator card will be used to help the hardware complete such tasks as object and facial recognition. The eight-lane, PCIe accelerator card is designed to be used with BrainChip Studio

ZTE achieves facial recognition breakthrough with deep learning

ZTE has achieved a breakthrough in deep learning and convolutional neural networks (CNN), working with Intel, which could enable new cloud applications performing picture search and matching, according to a Tuesday announcement. The companies say many others researching Internet search and artificial intelligence have been trying to reach the performance and accuracy benchmark. ZTE engineers in Nanjing City, China, used Intel’s mid-range Arria 10 FPGA (or field-programmable gate array) for a cloud inferencing application using a CNN algorithm, and reached

KnuEdge looking for speakers for heterogeneous neural networks event

As part of its effort to advance neural computing innovation and transform human-machine interaction, KnuEdge has opened a call for papers to be presented at its heterogeneous neural networks (HNN) workshop in April 2017. The event is expected to grow into a series, and is presented in partnership with the California Institute for Telecommunications and Information Technology (Calit2) at UC San Diego. KnuEdge is looking for academic papers on the benefits of sparse networks in theory and practice from fields

Chinese app developer releases facial recognition based lending platform

Chinese application developer PING AN TianXiaTong (PING AN) has released facial recognition technology that it will incorporate its wealth management solutions and other services in an effort to create the fastest lending platform in China, requiring only six minutes to complete a loan application. The PING AN technology is based on a complex neural network which allows computers to act like the human brain, thus achieving facial recognition with even greater accuracy than the human eye. The application also compensates

Russian photographer uses facial recognition in social media experiment

A recent project entitled ‘Your face is big data’ saw an art school student photograph people who happened to sit across from him on the subway and then he used FindFace, a facial recognition app that taps neural-network technology, to track them down on Russian social media site VK. The FindFace service was designed for users of the largest Russian social network “Vkontakte” and is based on face recognition technology developed by N-Tech.Lab. According to a report by PC World,