Cortex (CTXC)

1 Articles 2 Followers

Cortex (CTXC) is an Artificial Intelligence (AI) focused blockchain platform that was launched in 2018 by Chen Ziqi. The team’s mission has been described as trying to “create a truly decentralized AI autonomous system.” The project’s investors include Bitmain, Huobi Capital, OK Blockchain Capital and many others.

Essentially, the team wants to bring advanced machine learning models to the blockchain, whereupon developers can leverage it to offer more efficient and practical decentralized applications (DApps). There are two main features to the Cortex platform: the Smart AI contract and the Cortex Virtual Machine.

The Smart AI contracts are smart contracts in which machine learning algorithms can be written, which can be also be made to depend on other contracts. The Cortex Virtual Machine is the platform’s own virtual machine that can support inference instructions. 

The team has also said that a submission interface will be available, which they say will “will setup a bridge between computing power providers and algorithm providers for trading and collaboration.” Users can also upload strong machine learning models, which other users can use after having paid for it. In the future, users will also be able to purchase graphics cards, deep learning machines and cloud computing services using the CTXC token.

Given the relative newness of the project, it is still very much in the preliminary development and testing phase. Goals outlined for the first two quarters of 2019 include a testnet for AI smart contracts and preparation for the mainnet launch. The project also established a few partnerships, most notably with Chainlink.

State of the Art Machine-Learning Models On Blockchain

16 Apr 2021 1 minute read 0 comments TheKoolzo555

Currently there is quite a challenge in executing machine learning on conventional blockchains, due to the fact that the virtual machine is very inefficient when running nontrivial machine learning models. Cortex is an open-source proof-of-work block...