The goal of Cortex is to create a truly decentralized artificial intelligence autonomous system, providing the most advanced machine learning model on the blockchain, and users can use the smart contract on the cortex blockchain to infer the model. One of the goals of Cortex also includes implementing a machine learning platform that allows users to post tasks on the platform and submit AI dapps.

Project Highlights

Cortex proposes AI DApps, the concept of artificial intelligence distributed applications. The introduction of artificial intelligence into smart contracts makes the following applications possible:

1. Information service: personalized recommendation system, search engine, news writing service.
2. Finance: credit, smart investment consulting.
3. AI assistant: automatic question answering, industry knowledge map, speech synthesis, face attribute prediction.
4. Simulation environment: automatic driving, Go and other reinforcement learning applications.

Cortex allows users to write machine learning programs on the blockchain and submit some interactions that depend on other contracts. This is an original point of Cortex, but the other contracts it refers to are smart contracts on ETH or smart contracts on different chains, which is not clearly stated in the white paper. If it is the latter, it will be more difficult to achieve. At present, the cross-chain of smart contracts is far from this level.

Architecture:
Data layer: Cortex's blockchain does not store models and data, but stores hash values ​​of models and data. Key-value storage systems are chainless. When enough copies are propagated throughout the network, the Cortex blockchain can acquire new models and new data.

Network layer: Cortex will add an inference instruction to the smart contract, which makes it possible for the smart contract to support the use of the Cortex blockchain model during the interaction process.

Consensus Layer: Use PoW to further narrow the gap between CPU and GPU speedup.

Incentive layer: Cortex proposes an incentive mechanism for collective collaboration, allowing anyone to submit and optimize models in Cortex, and model contributors can be rewarded.

Contract layer: Cortex will add an inference instruction to smart contracts, which makes it possible to support the use of Cortex blockchain models in smart contracts.

Application layer, platform layer: Cortex's goal also includes implementing a machine learning platform that allows users to publish tasks on the platform, submit models, make inferences by calling smart contracts, and create their own AI DApps (artificial intelligence decentralized applications program).

Application scenario

Information service, financial service, artificial intelligence assistant, simulation environment, etc.