Arm and Baidu PaddlePaddle Build an AI Ecosystem Together
Arm and Baidu PaddlePaddle, China's premier open-source deep learning platform, are working together to accelerate AI-based IoT development opportunities in a brand new ecosystem.
By Sherry Gu, Director, Ecosystem, China Line of Business, Arm
What do you get when you combine China’s premier open-source deep learning platform with Arm? Rocket fuel for innovation.
Arm announced recently that it has partnered with Baidu PaddlePaddle, the deep learning platform, to accelerate development of transformative IoT solutions leveraging AI. The announcement came during the WAVE Summit 2022 for Deep Learning, where Arm and Arm China joined Baidu’s PaddlePaddle Hardware Eco-Partner Co-creation Program as a partner. The Eco-Partner Program specifically includes joint technology research and development, promotion and resource sharing, win-win ecological business opportunities and training enabling growth. Arm and Baidu will jointly invest resources in the next cooperation to create a software-hardware integrated artificial intelligence ecosystem.
Also at the summit, Baidu PaddlePaddle, together with several partners, launched the Hardware Ecological Co-creation Plan, hoping to use their respective technologies and market application experience to jointly build an efficient software-hardware integrated platform solution. Through the continuous upgrading of product technology and the continuous optimization of functional experience, the implementation of the AI industry will be accelerated and achieve ecological prosperity and a win-win situation.
PaddlePaddle is the first open-source deep-learning platform in China. Similar to AI frameworks such as Google’s TensorFlow and Facebook’s PyTorch, PaddlePaddle, which was open sourced in 2016, provides software developers of all skill levels with the tools, services, and resources they need to rapidly adopt and implement deep learning at scale.
For example, Jinlu Technology has leveraged PaddlePaddle to build a garbage-sorting robot – programmed with an object-detection model – that can discern a vast array of items using vision and AI. It also uses an image-segmentation model to find garbage and do things like detect the edge of a bottle and determine its center point. The model takes just half a second to recognize an image.
Building on a years-long partnership in other areas and looking to a future built on Arm, Arm and Baidu PaddlePaddle will deepen cooperation and create a software-hardware integrated ecosystem for deep learning. This combines the latest Arm Virtual Hardware with the PaddlePaddle platform to provide developers with a highly flexible and scalable IoT virtual development platform. At the same time, the companies jointly launched Model Zoo, an industrial-level model library that runs efficiently on the Arm IoT platform.
Baidu PaddlePaddle and Arm have a shared vision of a mobile hardware ecosystem, and therefore, the two parties have been in a long-term collaboration that has included Baidu PaddlePaddle joining the Arm AI Partner Program. Arm’s extensive AI ecosystem simplifies AI deployment on intelligent endpoint devices by providing best-in-class tools, algorithms and applications to businesses worldwide.
In the past few months, Arm Compute Library (ACL) team has deeply engaged with PaddlePaddle’s core R&D team to facilitate the improvement of overall performance on Arm Cortex-A CPU and Mali GPU on mobile and edge devices. The collaboration aims to enable better user experience when they use Arm-based hardware as back-end inference engine.
By creating and nurturing strategic alliances that empower its ecosystem to drive innovation, Arm focuses on helping its partners reach developers and decision-makers within their target markets.
The partnership is clearly a pivotal moment in the evolution of AI development in China. After more than 6 years of development, Baidu PaddlePaddle now ranks first in the comprehensive share of open-source deep learning frameworks in China. Nearly 5 million developers have created 560,000 models based on the PaddlePaddle open-source deep learning platform, serving 180,000 enterprises and institutions. PaddlePaddle helps developers quickly realize AI ideas, innovate AI applications, and serve as a basic platform to support more and more industries to realize industrial intelligent upgrades.
Pleased to join @docomo as an OREX partner, supporting overseas telecom operators implement Open RAN.
This will see OREX packages expand to Arm-based, power efficient Open RAN chipsets from @AmpereComputing, with @NEC & @Fujitsu_Global software: https://ssw.web.docomo.ne.jp/orex/en/topics/20240226_01.html #MWC24
🎮 More realistic, immersive gaming
👾 AI-based characters
🔋 Power savings for longer game-time
With the Immortalis GPUs and range of optimization technologies, the future of mobile gaming is here and it runs #onArm: https://newsroom.arm.com/mobile-gaming-techniques?utm_source=twitter&utm_medium=social-organic&utm_content=blog&utm_campaign=mk19_events_na_2023 #MWC24
Day one of #MWC24 showed us how sustainable mobile networks can fuel the growth of AI to deliver unparalleled user experiences.
We're pleased to work closely with our partner ecosystem to build this future #onArm Neoverse. Here's how: https://newsroom.arm.com/sustainable-cloud-infrastructure?utm_source=twitter&utm_medium=social-organic&utm_content=blog&utm_campaign=mk19_events_na_2023
We're joining forces with tech leaders and academic institutions as a member of the AI-RAN Alliance, announced today at #MWC24.🙌
Tying research together with testing and implementation, we'll explore how to use AI to advance RAN technology for 5G and 6G: https://ai-ran.org/news/industry-leaders-in-ai-and-wireless-form-ai-ran-alliance/