Story image

Toshiba develops a high-speed algorithm for deep learning processors

07 Nov 18

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.

Dimension Data nabs three Cisco partner awards
Cisco announced the awards, including APJ Partner of the Year, at a global awards reception during its annual partner conference.
WatchGuard’s eight (terrifying) 2019 security predictions
The next evolution of ransomware, escalating nation-state attacks, biometric hacking, Wi-Fi protocol security, and Die Hard fiction becomes reality.
Rimini Street hits NZ shores with new subsidiary
The third-party support provider for Oracle and SAP has opened a new Auckland-based office and appointed Sean Jones as NZ senior account executive.
Why the adoption of SAP is growing among SMEs
Small and medium scale enterprises are emerging as lucrative end users for SAP.
Exclusive: How the separation of Amazon and AWS could affect the cloud market
"Amazon Web Services is one of the rare companies that can be a market leader but remain ruthlessly innovative and agile."
HPE extends cloud-based AI tool InfoSight to servers
HPE asserts it is a big deal as the system can drive down operating costs, plug disruptive performance gaps, and free up time to allow IT staff to innovate.
Digital Realty opens new AU data centre – and announces another one
On the day that Digital Realty cut the ribbon for its new Sydney data centre, it revealed that it will soon begin developing another one.
A roadmap to AI project success
Five keys preparation tasks, and eight implementation elements to keep in mind when developing and implementing an AI service.