NeuroMesh: Spearheading the new era of AI with a distributed training protocol

April 09, 2024 07:20 AM PDT | By Invezz

NeuroMesh: Spearheading the new era of AI with a distributed training protocol — Image source: Invezz

Visionaries in AI: The team’s global ambition

The team behind NeuroMesh, composed of researchers and engineers from Oxford, NTU, PKU, THU, HKU, Google, and Meta, pioneers a democratic AI training process. This visionary approach addresses the limitations of centralized AI development by enabling GPU owners worldwide to contribute to a vast training network, empowering entities of all sizes to leverage this service for their training needs.

NeuroMesh transcends traditional AI by fostering collaboration. Their vision is to equip every developer and organization, regardless of location or resources, with the ability to train and utilize cutting-edge AI models. This aligns perfectly with the vision of AI pioneers like Yann LeCun, who advocate for a future powered by crowdsourced and distributed AI training.

A revolutionary design based on PCN

At the heart of NeuroMesh’s distributed training protocol lies the groundbreaking PCN (Predictive Coding Network) training algorithm – a true game-changer in this field. This approach empowers GPU owners worldwide to contribute their power, fostering a vast collaborative effort.

The PCN Training Algorithm: The magic behind NeuroMesh lies in the PCN training algorithm. Unlike traditional backpropagation (BP) methods, PCN enables fully local, parallel, and autonomous training. The team aims to create a vast network, where each node—representing a participating GPU—learns independently. PCN minimizes inter-layer communication, reducing data traffic and facilitating asynchronous training. Think of it as a symphony where each musician plays their part independently, yet contributes to a harmonious whole.

This cutting-edge model, inspired by recent advancements in neuroscience research pioneered by Oxford University, mimics the human brain’s localized learning approach. By storing error values and optimizing for a local target in each layer, it replicates the behavior of brain neurons. This allows NeuroMesh to define models that are much larger, with individual components that contribute to the same ultimate optimization objective for the whole network, just like the human brain where different stimuli are handled by different groups of neurons.

This biologically-inspired approach, combined with its inherent distribution capabilities, unlocks a new era of AI development.

A call to forge global partnerships

NeuroMesh invites partnerships globally, aiming to forge an AI future that everyone can participate in. Its protocol is the bedrock upon which a diverse ecosystem is being built. The ecosystem is designed to be dynamic, collaborative, and adaptable, ensuring that it can serve the AI model training needs of any size, from any industry.

Individuals, projects with GPU resources, and entities with training needs are all welcome to join this transformative initiative. For comprehensive details on NeuroMesh and to participate in this leading-edge endeavor, users can visit nmesh.io.

About NeuroMesh

NeuroMesh comprises researchers and engineers from esteemed institutions such as Oxford, NTU, PKU, THU, HKU, Google and Meta. By empowering developers and organizations to deploy robust AI models, NeuroMesh is cultivating an inclusive AI ecosystem, bridging the gaps between the demand of training large AI models and distributed GPUs worldwide.

For more information, users can visit NeuroMesh’s

Contact

CMO
Kenchia Lee
NeuroMesh
[email protected]
07746906341

The post NeuroMesh: Spearheading the new era of AI with a distributed training protocol appeared first on Invezz

Disclaimer

The content, including but not limited to any articles, news, quotes, information, data, text, reports, ratings, opinions, images, photos, graphics, graphs, charts, animations, and video (Content) is a service of Kalkine Media LLC., having Delaware File No. 4697309 (“Kalkine Media, we or us”) and is available for personal and non-commercial use only. The principal purpose of the Content is to educate and inform. The Content does not contain or imply any recommendation or opinion intended to influence your financial decisions and must not be relied upon by you as such. Some of the Content on this website may be sponsored/non-sponsored, as applicable, but is NOT a solicitation or recommendation to buy, sell or hold the stocks of the company(s) or engage in any investment activity under discussion. Kalkine Media is neither licensed nor qualified to provide investment advice through this platform. Users should make their own enquiries about any investments and Kalkine Media strongly suggests the users to seek advice from a financial adviser, stockbroker or other professional (including taxation and legal advice), as necessary. Kalkine Media hereby disclaims any and all the liabilities to any user for any direct, indirect, implied, punitive, special, incidental or other consequential damages arising from any use of the Content on this website, which is provided without warranties. The views expressed in the Content by the guests, if any, are their own and do not necessarily represent the views or opinions of Kalkine Media.
The content published on Kalkine Media also includes feeds sourced from third-party providers. Kalkine does not assert any ownership rights over the content provided by these third-party sources. The inclusion of such feeds on the Website is for informational purposes only. Kalkine does not guarantee the accuracy, completeness, or reliability of the content obtained from third-party feeds. Furthermore, Kalkine Media shall not be held liable for any errors, omissions, or inaccuracies in the content obtained from third-party feeds, nor for any damages or losses arising from the use of such content. Some of the images/music that may be used on this website are copyrighted to their respective owner(s). Kalkine Media does not claim ownership of any of the pictures/music displayed/used on this website unless stated otherwise. The images/music that may be used on this website are taken from various sources on the internet, including paid subscriptions or are believed to be in public domain. We have used reasonable efforts to accredit the source (public domain/CC0 status) to where it was found and indicated it, as necessary.
This disclaimer is subject to change without notice. Users are advised to review this disclaimer periodically for any updates or modifications.