- Raven Protocol is a decentralized and distributed deep-learning training protocol.
- It provides cost-efficient and faster training of deep neural networks.
- Raven Protocol facilitates dynamic allocation of nodes to the participating devices in the network. Hence it eliminates any added dependencies on the host nodes and substantially reduces the compute power required for in house computing.
- Raven Protocol builds a dynamic graph for vast number of small synchronous calculations that are required to train a model.
- Raven Protocol allows for sharing of resources from idle compute power of individual contributor’s devices.
- The notion of sharing idle computing power of individual contributor’s devices to facilitate training saves the high expense involved.
- Individual contributors are rewarded with Raven tokens for sharing their devices.
- The incentive process involves smart contracts in the Ethereum Blockchain.
- Thus Raven will be able to offer compute services at a far cheaper price than any provider in the market.
- Raven aims to help individuals and companies to fully exploit the potential of AI, economically.
To learn more about Raven please visit the websites below
