EpiK Protocol
Published in

EpiK Protocol

Use the “Marvel Universe” to Explain Knowledge Graph and Artificial Intelligence

If time has a direction, then everything should have a direction, and the accumulation of data will not always be chaotic. “Data”, something that has been mentioned and discussed for years, is about to have its key moment: the knowledge graph.

I hope that after reading this article, you will be able to answer the following questions.

  • What is the Knowledge Graph?
  • How do you briefly understand artificial intelligence?
  • What is the ultimate goal of EpiK and what is the path to get there?
  • How does EpiK achieve “broaden AI’s horizon”?

Basic Concepts and History of Knowledge Graph

The semantic network was introduced in 1956, before the concept of artificial intelligence, and even the Internet was still in its infancy. The semantic network is essentially a product of the development of logic and mechanical engineering, aiming at making machines understand information through semantic representations and solving it automatically based on knowledge.

There is no fundamental difference between a knowledge base and a knowledge graph, except that a knowledge graph is a more connected representation of knowledge. The core of the semantic network is the knowledge graph. Recently, Web 3.0 and AI are emerging again, but the knowledge, concepts, and applications of knowledge graph have not been well discussed and understood in the field of blockchain. The articles on the web are too technical and obscure to understand a concept or thing with a new concept that no one understands. So how to understand it? Let’s use the most familiar story as an analogy.

The Knowledge Graph of the Marvel Universe

Any knowledge, or any information called knowledge, can be vast and complex, but it is never completely disordered. Readers, please recall the story of the Marvel Universe. What can you think of?

The first thing that comes to mind is that the story takes place in modern times, but the storyline begins in World War II, spatially from the United States to the world to the universe, and then we think of the characters: tony stark (Iron Man), Captain America, the Hulk, Hawkeye, Thor, and so on.

Then a picture popped into my head.

Diagram of the relationship between the Marvel Universe, image from the web.

What’s on this diagram? It is the relationship between characters, but also the relationship between wars, between races, and between events that can be sorted out.

Our understanding of a story requires a framework and a set of rules, a timeline in our minds, multiple narrative scenarios, the progression of different events, and a variety of character relationships and war posture adjustments. From this seemingly vast amount of information, it is woven into a network of categories and becomes the “knowledge graph. This is how a mature human being thinks.

What if you were to ask a seven or eight-year-old the same question?

This child is also very smart and reads comic books. Ask him the same question and ask him to tell a comic book story. Kids say, “I know Captain America has a shield, I know Iron Man is cool, I know Spider-Man jumps around, and I even know that a purple-faced villain named Thanos is a good fighter, but not much else.

It can be argued that the level of cognition is rising from childhood to adulthood because the body of knowledge is becoming more sophisticated. If we compare our own childhood and adulthood, we can conclude that our level of intelligence is rising.

A Popular Understanding of Artificial Intelligence

There are many statements in the AI field to express the current state of AI, such as this famous assertion: “Strong AI is weaker than humans, weak AI is stronger than humans”. It sounds a bit like a tongue twister, but it’s actually quite easy to understand.

For example, an AI can easily know who the main characters are in the Marvel Universe, perhaps by retrieving whose name appears most often in a Marvel comic book. But in terms of the number of times a person’s name appears, this AI is so low-end that it is like a retrieval tool, belonging to the category of weak AI, which is certainly more accurate than a human’s answer — this is “weak AI is better than human”.

If both AI and human read that Captain America went back to the past to meet his beloved Peggy Carter (during World War II), and deduced the future direction of the story, the human might think more comprehensively than him, and then the human might come to the conclusion that Captain America might not go back to the present day according to the established story direction, but go back to the past after he and his buddies finished saving the world! The romance of spending the rest of your life with the woman you love, the romance of making decisions that go beyond basic logic, is relative to “strong artificial intelligence. Machines can’t tell, but intelligent humans can make relatively accurate judgments.

Captain America and Peggy Dance

Try to be a hero and stay with your lover.

The reason for this is that as humans continue to absorb information, they compile and organize it into a knowledge graph, and a perfect knowledge graph makes the thinking process more complete.

Machine intelligence is also developing, and many people realize that the efficiency and function of machines are not good enough in many fields, but they want machines to improve their cognitive level, so the theory and practice of knowledge graph are in progress. The most famous one is that Google added a “knowledge graph” to its search engine in 2012. In 2012, Google added 「Knowledge Vault」 to its search engine and proposed a knowledge fusion framework called “Knowledge Vault” to build large-scale knowledge graphs. Since then, other centralized companies have been refining and pondering this path of AI evolution.

The Ultimate Goal of EpiK and the Path to Get There

The ultimate goal of EpiK is to “build the eternal knowledge vault of mankind and broaden AI‘s horizon”.

Some people don’t understand why a distributed storage project says that it can broaden the horizon of AI? Let me explain in detail.

The knowledge graph is an important foundation for Web 3.0, and it’s what EpiK is trying to build.

EpiK is a distributed storage project, but it’s more than just a distributed storage project.

The point of distributed storage is to aggregate information, data, and to maximize the uptake of information through blockchain technology with the incentives of a token economy. Powerful centralized companies hold a lot of data, but two things are worth noting.

First, the ownership of the data is ambiguous and more often than not is a profit-making tool for giant corporations. Individuals who provide the data can not benefit from it.

Second, there are geo-barriers to a centralized network. Google built the framework of the Knowledge Graph in 2012, but the Knowledge Graph does not have much data on Chinese people for the simple reason that very few people in mainland China use Google. The same goes for Baidu, which doesn’t have much data on foreigners. Even if a data island grows big and becomes a giant, it is still an island, and there will always be bottlenecks in development.

Only with decentralized data collection can we maximize data collection and solve the problem of data entitlement and incentive. This is the first feature of EpiK as a distributed network: distributed storage.

Field Experts and Knowledge Graphs

If EpiK were just a storage project, it wouldn’t be worth the attention. The addition of field experts to the project is the finishing touch, giving the data soul. A field expert reviews the data and weaves the useful data into a knowledge graph. This data goes from chaos and disorder to vector space, i.e., to an ordered knowledge graph. Because of this, EpiK is able to speak for itself in terms of broadening the horizon of AI.

To summarize, the path of EpiK is as follows: a distributed storage as a vault on which a persistent knowledge graph is built, and eventually the knowledge graph is combined with AI to broaden AI’s horizon.

The analogy can be simplified as follows: distributed storage is a foundation, on which the knowledge graph builds a framework. And the materials in this framework are rich and orderly, and then AI enters to fill and decorate the building.

Can’t other storage projects do knowledge graphs?

Of course, you may also ask, now that there are several distributed storage projects. Couldn’t they follow up with knowledge graphs to serve AI as well? Of course there is some probability, but the probability is extremely small. It is very difficult to build a knowledge graph on top of the chaotic and useless data that has already been assembled as if you need to “blow a bunch of steel and rubber into the sky with a strong wind and then turn it into a car when it hits the ground”. The EpiK Protocol does this.


This article describes in layman’s terms the concept and history of the knowledge graph, the development and evolution of AI, the path and ultimate goal of EpiK, and the details of these concepts may differ slightly from the academic understanding, but it does not affect the understanding.

Epilogue: AI Cognitive Rhapsody

The knowledge vault that EpiK will build in the future will broaden the horizon of AI, so what AI will look like in the future? I don’t know. Assume that the classic quote from Spider-Man in the Marvel Universe has never even existed in human history. In the future, the EpiK project is well developed. The consensus is expanding, the knowledge graph is enriched, and the AI is broadening its cognition under the auspices of EpiK’s powerful knowledge graph, when you say to the AI, “Can you summarize the logic of the behavior of all comic book heroes in one sentence?”

The screen in front of you pops up with the words.

“With great power comes great responsibility”

Of course, this is also a joke. Perhaps with the EpiK Knowledge Graph and the evolution of AI, all of this will become a reality in the near future, and although the EpiK project is a small team with less than 100 people, it has big dreams and a clear path, and we hope that you will join us to realize the vision that is prominent on the official website. “Building the Eternal Knowledge Vault of Humanity and Broadening the Horizon of AI”.



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store