[ad_1]
London, United Kingdom, April 9, 2024, Chainwire
NeuroMesh (nmesh.io), a pioneer in artificial intelligence, announces the launch of its distributed AI training protocol, designed to revolutionize global access and collaboration in AI development. Leveraging DePIN's decentralized framework, NeuroMesh bridges the gaps between the demand for training large AI models and distributed GPUs. This initiative aims to promote inclusion in AI development and facilitate participation across different sectors and regions.
Visionaries in AI: The team's global ambition
The team behind NeuroMesh, made up of researchers and engineers from Oxford, NTU, PKU, THU, HKU, Google (NASDAQ:) and Meta (NASDAQ:), is pioneering a democratic AI training process. This visionary approach overcomes the limitations of centralized AI development by enabling GPU owners worldwide to contribute to a vast training network, allowing companies of all sizes to leverage this service for their training needs.
NeuroMesh goes beyond traditional AI by encouraging collaboration. Their vision is to give every developer and every organization, regardless of location or resources, the opportunity to train and use cutting-edge AI models. This fits perfectly with the vision of AI pioneers like Yann LeCun, who are committed to a future based on crowdsourcing and distributed AI training.
A revolutionary design based on PCN
At the heart of NeuroMesh's distributed training protocol is the groundbreaking PCN (Predictive Coding Network) training algorithm – a real game-changer in this area. This approach allows GPU owners worldwide to contribute their performance, driving tremendous collaboration.
The PCN Training Algorithm: The magic behind NeuroMesh lies in the PCN training algorithm. Unlike traditional backpropagation methods (BP (NYSE:)), PCN enables fully local, parallel and autonomous training. The team's goal is to create a massive network in which each node – representing a participating GPU – learns independently. PCN minimizes communication between layers, reduces data traffic, and facilitates asynchronous training. Think of it as a symphony, with each musician playing their part independently while still contributing to a harmonious whole.
Inspired by recent advances in neuroscience research at the University of Oxford, this cutting-edge model mimics the human brain's localized approach to learning. By storing error values and optimizing for a local target in each layer, the behavior of brain neurons is modeled. This allows NeuroMesh to define models that are much larger and whose individual components contribute to the same ultimate optimization goal for the entire network, just like the human brain, where different stimuli are processed by different groups of neurons.
This biologically inspired approach, combined with the inherent dissemination capabilities, ushers in a new era of AI development.
A call to build global partnerships
NeuroMesh invites global partnerships to create an AI future that everyone can participate in. Its protocol is the foundation on which a diverse ecosystem is built. The ecosystem is designed to be dynamic, collaborative and adaptable to ensure it can meet the training needs of AI models of any size and industry.
Individuals, projects with GPU resources, and companies with training needs are all welcome to join this transformative initiative. To get comprehensive information about NeuroMesh and get involved in this groundbreaking project, users can visit nmesh.io.
About NeuroMesh
NeuroMesh consists of researchers and engineers from renowned institutions such as Oxford, NTU, PKU, THU, HKU, Google and Meta. By empowering developers and organizations to deploy robust AI models, NeuroMesh cultivates an inclusive AI ecosystem and bridges the gaps between the demand for training large AI models and distributed GPUs worldwide.
For more information, users can visit NeuroMesh's Twitter | visit Telegram
ContactCMOKenya LeeNeuroMesh[email protected]07746906341
[ad_2]
Source link
