Baidu to Announce Kunlun AI Chip

The CPU is designed to perform edge computing on the device itself and cloud through data centers
06 July 2018   756

At the annual Beijing conference for developers Baidu Create AI, the company disclosed details about the Kunlun processor for computations related to artificial intelligence. In addition, the release of the platform Baidu Brain 3.0 with new "smart" services was announced.

Kunlun Announcement
Kunlun Announcement

According to the company representatives, the processor is designed to perform edge computing on the device itself and cloud through data centers. It is known about two models of chips: 818-300 for training algorithms and 818-100 for outputting the result.

It is assumed that Kunlun will be able to perform up to 260 ter-operations per second. The memory bandwidth is 512 GB per second. The company did not specify a specific release date.

In addition to Baidu, many technology companies work on their own chips for intelligent systems, for example, Alibaba or Tesla, and Intel and AMD have already presented specific solutions.

In the third version of the Baidu platform, developers added 50 new AI services, from computer vision systems and natural language processing to face recognition software. In addition, the update improves the semantic understanding of input data - models better capture the meaning of text, video and images.

DuerOS Presentation
DuerOS Presentation 

The company also upgraded the DuerOS service for conversational AI technologies to version 3.0. It now supports home devices - televisions, smart speakers or clocks, as well as mobile devices. Thus, developers can use their open SDK and API to create their own voice control systems.

Neural Network to Create Landscapes from Sketches

Nvidia created GauGAN model that uses generative-competitive neural networks to process segmented images and create beautiful landscapes from peoples' sketches
20 March 2019   150

At the GTC 2019 conference, NVIDIA presented a demo version of the GauGAN neural network, which can turn sketchy drawings into photorealistic images.

The GauGAN model, named after the famous artist Paul Gauguin, uses generative-competitive neural networks to process segmented images. The generator creates an image and transfers it to the discriminator trained in real photographs. He in turn pixel-by-pixel tells the generator what to fix and where.

Simply put, the principle of the neural network is similar to the coloring of the coloring, but instead of children's drawings, it produces beautiful landscapes. Its creators emphasize that it does not just glue pieces of images, but generates unique ones, like a real artist.

Among other things, the neural network is able to imitate the styles of various artists and change the times of the day and year in the image. It also generates realistic reflections on water surfaces, such as ponds and rivers.

So far, GauGAN is configured to work with landscapes, but the neural network architecture allows us to train it to create urban images as well. The source text of the report in PDF is available here.

GauGAN can be useful to both architects and city planners, and landscape designers with game developers. An AI that understands what the real world looks like will simplify the implementation of their ideas and help you quickly change them. Soon the neural network will be available on the AI ​​Playground.