Huawei’s AI Strategy and Product Portfolio

HC logo shutterstock_1083161909

Unveiled at Huawei Connect 2018, Huawei’s new vision is a call for an intelligent world powered by artificial intelligence (AI). This in turn has shaped our AI strategy – a strategy that’s predicated on ten factors that will pave the way for widespread AI adoption. These are:

  1. Faster model training
  2. Abundant and affordable computing power
  3. AI deployment and user privacy
  4. New algorithms
  5. AI automation
  6. Practical application
  7. Real-time, closed-loop system
  8. Multi-tech synergy
  9. Platform support
  10. Talent availability

In a series of Huawei Connect-themed posts, we’ll be looking at each of these factors in more detail, as well as Huawei’s AI product portfolio and industry applications. In this post I’ve outlined Huawei’s main areas of focus in the AI domain plus our full-stack, inclusive, and all-scenario AI product portfolio.

The situation today:

  • 10% of enterprise applications are on cloud.
  • 13% of available data is utilized by enterprises.
  • 5% of companies have experimented with AI, and less than 1% have deployed AI solutions.

Our predictions for 2025 – a pivotal year for technological maturity – are that:

  • 85% of enterprise applications are on cloud
  • 80% of available data will be utilized by enterprises
  • 86% of companies will have deployed AI
Source: GIV2025

How Do We Reach The Stage of Pervasive Intelligence?

As well as providing the digital infrastructure that enables AI, Huawei has a mature full-stack AI strategy comprising five key areas of focus:

  1. Invest in AI research to develop capabilities for machine learning in domains like computer vision, natural language processing, and decision and inference. Research areas include:
  • Data and power-efficiency that requires less overall data, computing resources, and power use
  • Security
  • Automation
  1. Build a full-stack AI portfolio:
  • Deliver abundant and affordable computing power
  • Provide an efficient and easy-to-use AI platform with full-pipeline services
  • Make the portfolio adaptive to all scenarios, both standalone and cooperative scenarios between cloud, edge, and device
  1. Develop an open ecosystem and talent: Collaborate with global academia, industries, and partners.
  2. Strengthen the existing portfolio: Introduce an AI mindset and techniques into existing products and solutions to create greater value and enhance competitive strengths.
  3. Drive operational efficiency at Huawei: Apply AI to massive volumes of routine business activities for better efficiency and quality.

Huawei’s AI portfolio: full-stack, all-scenario, and inclusive

First, let’s define what we mean by the above terms.

Full-stack: refers to the functionality of Huawei’s technology, which includes chips, chip enablement, a training and inference framework, and application enablement.

All scenarios: indicates the different deployment scenarios for AI, including public clouds, private clouds, edge computing, industrial IoT devices, and consumer devices.

Inclusive AI: refers to Huawei’s commitment to working with all stakeholders to turn AI into a practical reality, making it inclusive and available for every person, home, and organization.

 The Portfolio

  • Ascend AI IP and chip series: the base layer of Huawei’s AI product portfolio, and the world’s first AI IP and chip series that serves all scenarios, providing optimal TeraOPS per watt. The Ascend series delivers excellent performance per watt in every scenario, including minimum energy consumption or maximum computing power in data centers. Unified architecture ensures easy deployment, migration, and interconnect AI applications across different scenarios.

The Ascend 910 and Ascend 310 chips mark Huawei’s leading AI capabilities at the chip level – the bottom layer of the stack. These chips will help greatly accelerate AI adoption in all industries.

  • CANN (Compute Architecture for Neural Networks): A chip operators library and highly automated operators development toolkit.
  • MindSpore: A unified training and inference framework for device, edge, and cloud (both standalone and cooperative).
  • Application enablement: Full-pipeline services (ModelArts), hierarchical APIs, and pre-integrated solutions.
  • The full-stack, all-scenario AI portfolio is designed to provide powerful support for Huawei Cloud EI – an AI service platform that will enable abundant computing resources for enterprises and governments – and HiAI, Huawei’s AI engine for smart devices. HiAI is one of the world’s first mobile AI computing platforms with a dedicated Neural-network Processing Unit (NPU). With HiAI APIs, developers can focus on developing new AI applications.
  • SoftCOM AI is our target architecture for self-driving telecom networks covering device and cloud infrastructure; network management and the control center; and network O&M. For O&M, SoftCOM can deliver Predictive Maintenance (PdM) capabilities whereby AI calculates the probability of device failure and schedules targeted maintenance before a fault occurs.

A Global Outlook

The path to the digital economy reached a tipping point in 2018. This year, Huawei’s Global Connectivity Index found that industries are embedding AI in five enabling technologies – Broadband, Data Centers, Cloud, Big Data and IoT – and transforming traditional connectivity into Intelligent Connectivity to unleash innovation and economic opportunities on a scale previously unimagined. Whole new business models, products, processes, and services are emerging every day, accelerating growth towards a digital economy worth upwards of US$23 trillion by 2025, and throwing open the gates to a new economic growth cycle.

Intelligent connectivity is one of the keys that will Activate Intelligence for every person, home, and organization.

One current issue affecting the industry is the availability of skilled AI talent – specifically the lack thereof. For companies to move from strategy to deployment, the shortage of AI talent needs to be addressed as it’s a key preconditions for AI readiness (as shown below).

GCI

Source: Huawei Global Connectivity Index 2018

Click on the link for an analysis by tech influencer and Huawei KOL Ronald van Loon covering how Huawei plans to help cultivate AI talent.

Coupled with E2E solutions for digital infrastructure and AI, Huawei is ready to Activate Intelligence for individuals, homes, and organizations for the betterment of everyone.

Further Reading:

Global Connectivity Index 2018: Huawei’s annual GCI report ranks nations according to the maturity of their digital infrastructure and assesses the wider economic impact. This year’s report focuses on Intelligent Connectivity, i.e., the application of AI into digital infrastructure.

GIV2025: This year’s Global Industry Vision provides direction for industries to ramp up their pace of development, explores the technological foundations that will enable a diverse ICT industry ecosystem, and gives the predictions for the intelligent world in 2025.

5 Thoughts

  1. Evan from Woodside. Sound’s exciting, It is said Brainchip are streets ahead in development with spiking neuromorphic network.
    What’s your thoughts?

Leave your comments

This site uses Akismet to reduce spam. Learn how your comment data is processed.