The speed of innovation within the machine learning ecosystem is faster than ever before, and shows no signs of slowing with new architectures regularly hitting the headlines. While such innovations are fantastic in driving performance and functionality, the sheer diversity of hardware and software solutions poses a challenge for developers hoping to support the latest advancements, while also reaching the widest audience.

Systems built for neural networks and machine learning typically base their performance on custom ML processors, CPUs, GPUs or FPGAs. This is an area in which Khronos is a recognized expert, being the home of some of the world’s leading standards for heterogeneous compute acceleration for embedded devices, desktops, and large HPC servers.

Machine Learning Standards from Khronos

Low-level, low-overhead cross-platform GPU API.	Low-level parallel programming of heterogeneous processors.	Single source C++ abstraction layer for heterogeneous processors.	Computer vision acceleration API with ML extension and tensor objects.	Exchange Format for connecting trained networks with inference engines.
Learn More	Learn More	Learn More	Learn More	Learn More

About the Khronos Machine Learning Forum

The Khronos Group warmly welcomes members of the ML community with an interest in training and inferencing acceleration to join the Khronos ML Forum. The forum was launched in April 2022 and provides all who join the following opportunities:

• Help guide the evolution of existing API standards and extensions to enhance their support for ML acceleration.
• Discuss the challenges of developing and deploying accelerated cross-platform ML systems.
• Share real-world use cases and build industry consensus on the key requirements for future ML acceleration standards.
• Provide feedback to Khronos members and API implementers to encourage consistent cross-platform ML support.
• Participate in webinars focused on achieving ML acceleration using Khronos standards.
• Network with peers in the machine learning community who share a keen interest in hardware acceleration.
• Attend regular Khronos ML Forum Summit meetings.

Features of the Machine Learning Forum

• Open to participants from any company, university, consortium, open-source participant, or individual working on machine learning acceleration.
• Free to join
• Open - No NDA required
• IP Free Discussions - High level requirements will be discussed but no detailed design discussions.
• An equal voice for all participants
• A dedicated Slack workspace
• Email distribution list for announcements, meeting invites and discussion
• Organised and hosted by The Khronos Group

Forum Meetings and Knowledge Sharing Webinars

All forum members will be invited to attend regular forum meetings where group members can present the key issues in machine learning acceleration and interoperability, and discuss possible solutions with each other and Khronos members. Khronos also realises that ML developers are often unaware or unclear on the best way to use the features and extensions that already exist in the current generation of Khronos APIs and associated extensions. To help address this issue we are running a series of knowledge sharing webinars for ML Forum members where a panel of experts will share their experience using Khronos API's in machine learning, discuss best practice, and answer questions from the audience.

* Provisional Dates