Chainge Technology Salon — Chain / AI Blockchain + Artificial Intelligence

  1. Chainge Technology Salon #6 — Kazuaki Ishiguro on Why Connectome Uses Token Curated Registry

The abbreviation TCR played a central role when Kazuaki Ishiguro, Regional Head of EEA and Cheif Blockchain Architect at Couger, took to the stage at the 11th Technology Salon “Chain” — Chain/ AI.

The event was held at the Shanghai Jiaoda Huigu Training Centre on 16 September 2018. It was organized by Blockchain EXE × 8BTC together with TomatoWallet, Bytom and Cortex.

“Chainge” is an offline event started by 8 BTC — the most influential independent platform for bitcoin, blockchain, and cryptocurrency news in China — to focus on blockchain’s hotspot discussion and communication between industries and companies. By hosting offline meetups, hackathons, blockchain summits, the organisers hope to help industry experts create high quality content, collaboration and promote the dissemination of knowledge and innovation in the field of blockchain.

Ishiguro, who is also a contributor for Vyper and Bigchain DB, started his presentation with a runthrough of Artificial Intelligence’s (AI) history from the 1950’s, through its first breakthrough with machine learning in early 80’s up to today.

“Its recent breakthrough is deep learning. What AI actually does is comprehension and decision making,” Ishiguro said.

Comprehension includes machine learning and deep learning. Decision making process are applied in appearance and personality. A central problem with AI today is the so-called black-box AI, which means systems where no one knows what is happening inside — other than potentially the one who created it. In comprehension AI, a lot of data and resources are required. These causes silo and black-box phenomenon, Ishiguro explained. There are also people have pointed out that AI is not reliable after the incident of Amazon Alexa or Google Home bug. Hence, there are some suggested that we need safe space to use AI but don’t know how to fix it.

Couger’s answer has been building a product called GeneFlow, which decentralizes deep learning. Distributing AI nodes, it adds in worker nodes under AI node. The Geneflow solution provides data integrity checks, algorithm failure checks, and long term effect checks.

“The comprehension side is safe as we applied blockchain technology. So now how are we going to solve the problem for decision making such as the black-box? Inside the decision making process, how it works is that if AI receives input A (which can be image or temperature,etc) then it will detect some information, then perform some action. In today’s AI, the process is hidden and what we see is just the output,” Ishiguro explained..

He and Couger believe that users should be able to see the process. Therefore, they created smart space and Token Curated Registry (TCR) use cases.

“In smart space use case, we make the AI smarter and able to detect what is going on and act based on the input. For example, in the situation where there are more than three persons seated in the meeting room, and not all of them have drinks, this is an input for the AI that will then find somebody in the office and ask them to bring the water to the meeting room. This will be stored in the smart contract so that everyone can see it,” Ishiguro said.

TCR is smart contract-based and uses Ocean Protocol, which is a data marketplace. Via Ocean Protocol’s datasets, Connectome creates new AI model.

To understand how it can operate in the AI, Couger held an internal hackathon, and created a new transaction using smart contract. The company is trying to create the best virtual human customer service and have created the token registry in the Connectome market, where all developers can create their own AI or sell algorithms and other people can vote for the best AI.