Harness the True Potential of Blockchain and Artificial Intelligence with AI 3.0
Harness the True Potential of Blockchain and Artificial Intelligence with AI 3.0
Find out what Grid Agents is and how this platform can allow Machine to Machine Interaction (M2M) to become a reality.
Join the DZone community and get the full member experience.Join For Free
The most visionary programmers today dream of what a robot could do, just like their counterparts in 1976 dreamed of what personal computers could do. Read more on MistyRobotics.com and enter to win your own Misty.
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) to become a reality.
The Grid Agents Platform
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 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 Agent 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 Agent.
Grid Agent 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:
- Common agreements
- Operationality in uncertain systems.
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 systems 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 that 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 into the tech stack. After researching on Hashgraph technology, I became 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."
Hashgraph Makes the Difference
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:
- Extremely fast transaction: formal result in real time.
- Fairness: proven by rigid mathematical models and timestamps. It will provide a fairness that individuals cannot alter.
- Security: bank-level security (Asynchronous Byzantine Fault Tolerant) prevents unwanted transactions from forming or achieving consensus.
- Unique: Hashgraph uses virtual voting and gossiping to reach a consensus instead of POW or POS, which our results show is quite effective.
- Based on the experienced team and investors behind Hashgraph, and its agile organization, we believe that Hashgraph will be the rule maker in blockchain industry.
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 for a few projects that include:
- Edge Computing emphasizing software-defined network security.
- Secure electricity system and energy exchange.
- Military intelligence (DOD).
- Supply chain operations with Internet security.
- Medical supply chain, with identity management and tracing.
Only time will tell, but I am glad to see the possibilities that 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.
Published at DZone with permission of Leona Zhang . See the original article here.
Opinions expressed by DZone contributors are their own.