Artificial Intelligence | News | Insights | AiThority
[bsfp-cryptocurrency style=”widget-18″ align=”marquee” columns=”6″ coins=”selected” coins-count=”6″ coins-selected=”BTC,ETH,XRP,LTC,EOS,ADA,XLM,NEO,LTC,EOS,XEM,DASH,USDT,BNB,QTUM,XVG,ONT,ZEC,STEEM” currency=”USD” title=”Cryptocurrency Widget” show_title=”0″ icon=”” scheme=”light” bs-show-desktop=”1″ bs-show-tablet=”1″ bs-show-phone=”1″ custom-css-class=”” custom-id=”” css=”.vc_custom_1523079266073{margin-bottom: 0px !important;padding-top: 0px !important;padding-bottom: 0px !important;}”]

SPEC Establishes Technical Committee on ML, Chaired by Inspur


Inspur, a leading datacenter, cloud computing and AI computing total solutions provider, hosts its annual Inspur Partner Forum (IPF). At the forum, the authoritative international evaluation organization SPEC announced the formal establishment of the Technical Committee on Machine Learning (SPEC ML), with Inspur’s representative as the inaugural Chair and Intel’s representative as the Secretary General. The committee will work on the development of Machine Learning Test Specification and related benchmarking. The SPEC ML Technical Committee consists of 12 members, namely Inspur, Intel, Alibaba, AMD, ARM, HPE, IBM, Indiana University, NetApp, NVIDIA, Oracle, and the University of Wuerzburg.

Read More: Aptiv/Audi Receives Innovation Partnership Award for Automated Driving Satellite Compute Platform

The mission of the ML Technical Committee is to establish and implement a set of unified and credible benchmarks for ML. Machine Learning is at the core of artificial intelligence, and its rapid development in recent years has driven the exponential growth of the overall AI industry. Computing infrastructure vendors have introduced a large number of solutions following the rapid development of accelerators like FPGAs and GPUs. In the scenarios of offline training and online inference, the performance difference between different hardware and computing frameworks is huge even for similar applications. The lack of benchmarks and methods for effective performance measurement has made it difficult for Machine Learning users, such as companies and research institutions, to make practical choices. In order to choose suitable solutions for themselves, users need an objective and impartial third-party evaluation benchmark.

Read More: Drones Officially Allowed to Deliver a Variety of Items in Australia’s Canberra

Peter Peng, Vice President of Inspur Group, believes that a unified test benchmark not only makes it easier for users to make choices, but also helps to break down the barriers between different technical camps and promote healthy competition in the ML industry. With a unified benchmark in place, different accelerators and computing frameworks can compete based on performance, which will promote enterprise innovation and ecological integration. At the same time, faster ML systems will tap the potential of the ML industry and even the entire AI industry more quickly.

Members of the SPEC ML Technical Committee include leading companies and scientific research institutions from various segments of the industry, such as chips, rack system, frameworks and applications, representing the most advanced technological forces in ML at the moment. For example, Alibaba has repeatedly ranked first in technical competitions in the ML field, and Inspur currently holds more than 50% of China’s AI infrastructure market share. Major enterprises of this nature are the key catalysts for the development of ML technology and applications. The composition of the ML Technical Committee fundamentally ensures the fairness and effectiveness of the ML performance test benchmark. When the specification is released, users will be provided with a feasible reference when choosing between different ML systems.

Read More: The Artificial Intelligence Week

Leave A Reply

Your email address will not be published.