I started working on AI and machine learning research in the early 90s. Since that time, I have attempted process prediction, coordination, and control with neural network and pattern matching technology. In 2003, I started a software project called Grid Agents. The Grid Agents platform uses a multi-agent system which allows nodes’ (server, device and other related things) access to the intelligent software. This intelligent software can self-configure, coordinate by passing messages, and can finish complex tasks using AI and machine learning. By using agent-systems to represent machines and to optimize by buying and selling, Grid Agents represents one of the earliest applications of the machine economy. In short, Grid Agents platform allows Machine to Machine interaction (M2M) become a reality.
An agent is a computing system. It can operate in a dynamic and unpredictable multi-agent system with flexible, distributed and self-determined operations. For example, in 2008, we worked with ConEdison to model the electric grid of entire Manhattan. We built Grid Agents and virtualized many aspects of automating the electricity delivery. We installed numerous Grid Edge modules (such as transformers, switches, inverters, surge protectors, distributed power grids, and power systems in buildings). We successfully virtualized many smart grid modules. This system automatically provides advance power outage notification to building owners if it detects any overheated underground transformer in the area. We were also able to trigger automatic shutdowns before a transformer would reach a critical condition. This prevented frequent transformer replacements.
Grid Agents can function as Edge Computing nodes and can self-coordinate several hundred smart grid modules. These functions can replace almost all the physical aspects of the system. That is why ConEdison is planning to include Grid Agents in all its future 3G systems. These days, Machine to Machine (M2M) systems such as smart cities, software-defined infrastructure, and mobile devices are rapidly employing Grid Agents.
Grid Agents and other generic multi-agent systems provide a powerful module to represent complex dynamic real environmental conditions. Yet we are omitting some important things from our introduction. These things have become a reality because of blockchain and distributed networks. They include:
Dr. Maxim Orlovsky pointed out:
"Blockchain will give multi-agent system AIs a way to evolve its sapience, just like what written language provided humanity."
Maxim also pointed out, what blockchains bring to multi-agent systems is a way to do consensus algorithm in the analysis. The consensus algorithm will provide uniform agreement about the state of things among all nodes in the system. Moreover, blockchains create an eternal memory, which makes the prediction in multi-agent system progressively better. We call blockchains cross multi-agent system AI 3.0. They will combine all the advantages of AI technologies, machine learning, and blockchains for the past 30 years, thus resulting in an emerging Industry 4.0. It will connect several billion devices with smart adjustments on Edge Computing.
These days, traditional blockchains, such as those behind Bitcoin are in vogue. However, traditional blockchain operations face difficulty in scaling up as the cost of mining, transaction charges, and the energy costs become unsustainably high.
Then there are no-blockchain cryptocurrencies, which attempt to mitigate all the scaling problems which come with Directed Acyclic Graphs (DAG). Examples include DAG Coin, the IOTA (the Internet of Things), Byte Balls and other emerging technologies. Many of these have found their suited use cases. However, as per my analysis, none of these meet my basic needs, such as scalability, speed, storage, error tolerance, network security, AI integration, and communication.
After spending a year, searching for a perfect basic blockchain to build my new software platform (which is sponsored by DOE and targeted on important physical network system tasks), I met the founders of Swirlds, Mance Harmon and Leemon Baird.
Mance told me that their team has a strong interest in AI, machine learning, and the recently emerging blockchain technologies. I also learned that Leemon had already developed a consensus algorithm based on distributed accounting. All this along with their individual accomplishments in the field made me strongly inclined towards their team.
Leemon Baird got his Ph.D. in computer science at Carnegie Mellon. He's also been the force behind many start-ups. He has also sold some companies in the past. Mance Harmon held a master in CS and had an important role at Ping Identity. As I wanted to learn more about the technology behind their work, I dug deep in the tech stack. After researching on Hashgraph technology, I became strongly interested in Hashgraph and joined the Hashgraph Team. The beauty of Hashgraph is in its performance and its simplicity.
As Leonardo da Vinci said, "Simplicity is complexity's final form."
As we learned from the Swirlds' team, Hashgraph is the future of the Internet and distributed technology. Hashgraph is designed to be a uniform data structure that replaces blockchains. As described on their website, a serverless distributed platform built on Hashgraph will give the following characteristics to a blockchain:
However, Hashgraph is currently a private authorized network. Yet, we are positioning it as a commercially viable solution for things such as micropayments, distributed money market, real-time coordination, distributed MMOs and much more. Currently, there is no cryptocurrency based on Hashgraph, but I can imagine cryptocurrency ecosystem will be a part of Hashgraph's future. I believe Hashgraph is a technology that will explode exponentially, and I have a plan to use Hashgraph in the initial licensing network. In the future, I want to be able to use Hashgraph on public networks. Currently, we are searching a few projects which include:
Only time will tell, but I am glad to see the possibility Hashgraph can provide to drive the economy. With Hashgraph, we truly have a technology that fills all the blanks between us and Industry 4.0.
Artificial Intelligence, Blockchain, and M2M interactions hold massive business value, but not without their challenges. Hashgraph provides an ideal approach to overcome these challenges and embrace emerging technologies. Hashgraph, holds the potential to be the future of distributed technology by combining all the advantages of AI, machine learning, and blockchain.
By David Allen Cohen
Alibaba Cloud RDS PostgreSQL – Time Series Data Optimization
2,599 posts | 762 followers
FollowUday Tank - November 15, 2022
Alibaba Clouder - June 12, 2018
Alibaba Cloud Community - July 27, 2023
Amuthan Nallathambi - July 12, 2024
PM - C2C_Yuan - August 14, 2023
Alibaba Cloud Community - January 30, 2024
2,599 posts | 762 followers
Follow
LedgerDB
A ledger database that provides powerful data audit capabilities.
Learn More
Offline Visual Intelligence Software Packages
Offline SDKs for visual production, such as image segmentation, video segmentation, and character recognition, based on deep learning technologies developed by Alibaba Cloud.
Learn More
Blockchain as a Service
BaaS provides an enterprise-level platform service based on leading blockchain technologies, which helps you build a trusted cloud infrastructure.
Learn More
Platform For AI
A platform that provides enterprise-level data modeling services based on machine learning algorithms to quickly meet your needs for data-driven operations.
Learn MoreMore Posts by Alibaba Clouder
