itb-nz logo
Story image

Toshiba develops a high-speed algorithm for deep learning processors

07 Nov 2018

Toshiba Memory Corporation announced the development of a high-speed and high-energy-efficiency algorithm and hardware architecture for deep learning processing with fewer degradations of recognition accuracy. 

The new processor for deep learning implemented on an FPGA achieves four times energy efficiency compared to conventional ones. 

Deep learning calculations generally require large amounts of multiply-accumulate (MAC) operations, and it has resulted in issues of long calculation time and large energy consumption. 

Although techniques reducing the number of bits to represent parameters (bit precision) have been proposed to reduce the total calculation amount, one of proposed algorithm reduces the bit precision down to one or two bit, those techniques cause degraded recognition accuracy. 

Toshiba Memory developed the new algorithm reducing MAC operations by optimising the bit precision of MAC operations for individual filters in each layer of a neural network. 

By using the new algorithm, the MAC operations can be reduced with less degradation of recognition accuracy.

Furthermore, Toshiba Memory developed a new hardware architecture, called the bit-parallel method, which is suitable for MAC operations with different bit precision. 

This method divides each various bit precision into a bit one by one and can execute 1-bit operation in numerous MAC units in parallel. 

It significantly improves the utilisation efficiency of the MAC units in the processor compared to conventional MAC architectures that execute in series.

Toshiba Memory implemented ResNet50, a deep neural network, on an FPGA using the various bit precision and bit-parallel MAC architecture. 

In the case of image recognition for the image dataset of ImageNet, the above technique supports that both operation time and energy consumption for recognising image data are reduced to 25 % with less recognition accuracy degradation, compared to conventional method.

Artificial intelligence (AI) is forecasted to be implemented in various devices. The developed high-speed and low-energy-consumption techniques for deep-learning processors are expected to be utilised for various edge devices like smartphones and HMDs and data centres which require low energy consumption. 

High-performance processors like GPU are important devices for high-speed operation of AI. 

Memories and storages are also one of the most important devices for AI which inevitably use big data. 

Toshiba Memory Corporation is continuously focusing on research and development of AI technologies as well as innovating memories and storages to lead data-oriented computing.

Link image
How to prioritise metrics as an e-commerce CTO
E-commerce technology leaders need to track, analyze, and act on large volumes of business and system performance data. Danny Miles, the CTO of Dollar Shave Club, shares a powerful framework for thinking about and prioritizing e-commerce metricsMore
Link image
Revealed: Which vendor leads the world in content services
The right content services platform offers a range of capabilities – from capture to collaboration to case management – as well as the industry expertise and customer support to help you drive your digital transformation initiatives. Download the report today.More
Story image
Cyber-attacks thrust IT compliance to the top of the business agenda
If an organisation is running on the cloud, its ecosystem has to be compliant with industry standards and frameworks. Here are some tips on how to ensure compliance processes are fit for purpose.More
Link image
Forget outages - here's how cloud-scale server monitoring can save your business
Whether you run Linux, Windows, Azure, AWS, or your internal network, server monitoring is critical to keeping your platforms in top condition.More
Story image
Internet outages drastically increased during COVID-19 lockdowns, report finds
Global internet disruptions increased 63% in March, with internet service providers hit the hardest. This is according to the 2020 Internet Performance Report from ThousandEyes, the internet and cloud intelligence company.More
Link image
Test before you leap: 5G emulation software that fits every use case
5G network emulation software combines innovation across all stages of 5G development using the same hardware and common software tools. More