WiMi Hologram Cloud Inc, a leading global Hologram Augmented Reality Technology provider, announced that the Proof of Learning (PoLe) blockchain consensus mechanism can enhance the training efficiency and data reliability of neural networks, which will inject a brand new impetus into the development of artificial intelligence.
PoLe is a novel blockchain consensus mechanism that uses consensus computation for the optimization of neural networks. In this design, training and testing data will be published to the entire blockchain network, and consensus nodes train neural network models on the data as proof of learning. PoLe is designed to motivate nodes to participate more actively in the training process by distributing rewards by verifying their learning outcomes in the neural network training process. During neural network training, nodes can prove their learning outcomes by submitting training results and model performance. If a node’s training results and model performance are verified as valid and excellent, then the node can receive a certain reward, and this reward mechanism can motivate nodes to participate more actively in the training process and improve the training efficiency and performance of the neural network.
The traditional neural network training process usually requires a large amount of computational resources and time, and data security and privacy protection during the training process is also an important challenge. PoLe consensus mechanism provides a new solution by combining blockchain technology with neural network training. The training and testing data will be released to the whole blockchain network, and will be processed and optimized through the PoLe Consensus Mechanism, and the final results obtained are then fed back into the neural network, which will realize the efficient training of neural networks. In addition, the non-tamperable characteristics of the blockchain also ensure the authenticity and integrity of the data in the training process, providing a highly reliable training foundation for neural network models, effectively combating data manipulation and fraud, and enhancing data credibility.
PoLe subverts the traditional Proof of Work (PoW) and Proof of Stake (PoS), and instead focuses on utilizing the learning outcomes of the machine learning process as the basis for reaching consensus. In short, PoLe encourages nodes in the network to participate in the training of neural network models by contributing computational resources, which greatly improves the efficiency of energy usage by directing computing power to meaningful neural network training tasks. From there, new blocks are verified and added to the blockchain. This mechanism not only optimizes resource allocation, but also significantly improves the speed and quality of model training. PoLe encourages more participants to contribute computational resources and data, forming a self-reinforcing ecosystem that accelerates model and technology innovation.
Additionally, the core component of PoLe is an SML that prevents consensus nodes from cheating and can be implemented directly as a linear neural network layer. When consensus is reached on the blockchain network, a new block is appended to the blockchain, resulting in a more stable block generation rate. By building consensus on the blockchain platform, PoLe not only enhances the transparency and security of data, but also greatly improves the efficiency and accuracy of neural network model training, which brings brand-new development opportunities for the AI field, and it will have extensive applications in industries that require high data security and model accuracy, such as healthcare, finance, and autonomous driving. For example, in the field of healthcare, through the blockchain network established by PoLe, different medical institutions can securely share anonymized patient data and jointly train disease diagnostic models, which protects privacy and improves diagnostic accuracy.
WiMi‘s PoLe blockchain consensus mechanism is not only a technology innovation, but also a far-reaching reflection on the future mode of building an intelligent society, which not only provides a new way for neural network training with high efficiency, security and synergy, but also opens up a broad space for cross-disciplinary technology integration and application expansion. With the continuous maturation of technology and the gradual improvement of ecology, PoLe is expected to become a key bridge connecting the real world and the intelligent future, leading us into a more intelligent and trustworthy era.
SOURCE: PRNewswire