Japanese

School of Informatics, Nagoya University

Special interview

Special interview

Let's find the surprise created by the artificial world

Reiji Suzuki
Associate Professor

【Currently】Graduate School of Information Science
【Planned lead】Department of Natural Informatics, School of Informatics

Reiji Suzuki 1

Emergent phenomena that creates evolution of virtual life forms

-What kind of research are you doing?

My research is the understanding of existence phenomena and applications of artificial life by creating groups of human society and life forms in artificial worlds on computers. For example, I investigate how diverse life forms are created from the interspecies relationship of predator and prey. I create an artificial world where block type virtual life forms - affectionately named "Jitabatan" (Wrigglers) - live in three-dimensional space. In this artificial world, there are predatory life forms called "Tabetan" (Eaters) who attack the Wrigglers. In the beginning, the Wrigglers are quickly devoured. However, elements were introduced so that they would evolve in this world, elements being differences in their shape and movement, differences being things like affecting their ease of survival and ability to transfer this to their offspring. I then investigate what type of life forms are created after repetitive generation shifts.

drawing 1

After actual evolution of several thousand generations, life forms are created that have a new shape and function not present at the beginning such as being able to wriggle blocks to move and escape predators. When interactions occur in their main composition and new functions and characteristics are born from the accumulation of small changes as in evolution this is called "Emergent phenomena". In particular, we call this a "surprise" and I believe it is an important key to understanding existence. It is interesting when life forms are born that protect vital points in a large block or various life forms are born that have different patterns each time I perform the experiment. There are many methods of resolving a problem and life forms can find new answers through the process of evolution which can be called emergence. I believe this is probably one of the reasons that creates are so diverse.

Surprised by emergent phenomena, the ability to discover various scenarios

-What knowledge is necessary to make the content of your research more tangible?

That would be knowledge to ask interesting questions about existence phenomena, particularly evolution. Also, the technology to create models for existence phenomena on computers, using Wrigglers as an example, that would be things like "genetic algorithms" that express the process of evolution. However, I think the most important is to feel that emergent phenomena is interesting. Emergent phenomena exists in human society as well as groups of life forms. For example, I am creating and investigating an artificial world to see what kind of conditions are necessary to have "cooperative" behavior spread in human society. When looking from the viewpoint of emergent phenomena, there are common points in both the mechanism and structure of human society and groups of life forms. We are attempting to understand these features from the viewpoint of information. Therefore, emergent phenomena can be found in various scenarios and it gives us the ability to survey something from an advantageous angle. What I mean is it is necessary to have the ability to see nature and society as a system.

Reiji Suzuki 2

Designing a soft robot with an evolution algorithm!?

-In what kind of situation can we see where your research field is useful?

As an example, evolution algorithms are also used in part when designing the front carriage of the bullet train or designing the wings on aircraft. Robots are one of the applications that we think about. We usually think of robots as hard objects but the use of soft materials such as rubber is also being considered. For example, walking robots created from parts that expand and contract like muscles and parts that do not move like bones are being created by 3D printers. It might not be easy to equip such robots with good functions or movement even though we are able to print their shape freely. In such a case, we can evolve robot groups in an artificial world to emerge them with the ability to walk well after hundreds of generations after they initially fail to walk reliably after several generations. I think it would be interesting if we came closer to "using an evolution algorithm to enable robots with the ability to move in a surprising way!"

drawing 2

Changes in values for computers and life forms. I glimpsed the "future" that is here now

-Why have you set your sights on this research?

From childhood I liked computers and also did some programming. At that time, I thought of computers as something that could be used to take on tedious and complicated work. However, I encountered artificial life at University and was greatly attracted to the idea of using computers to create something that would exceed my own imagination. In fact, I was not really interested in life forms until then and I awakened to the truth that life forms were interesting from the viewpoint of emergence and information. Now I have actually started to also examine and research bird songs. Evolution algorithms are being applied now and over a wide range in society. The demand to understand existence has increased due to interest in technology relating to information and when thinking about this, I feel that I am at a position where I can glimpse the future.

Reiji Suzuki 3

Look, touch and combine. Discover your own value

-Do you have a message for students wishing to join the School of Informatics?

You can now easily acquire knowledge alone but it is important to have the "ability to discover" things that are really interesting to you. I think you will change based on how you place yourself in situations that allow you to look, touch and experience many fields. It is important to have the ability to combine various knowledge and discover value and fun that is distinct to you. At the school of informatics and science there are people from various fields researching at the cutting edge using "information" as the key. It is an ideal place where you can directly come into contact with things you haven't focused on until now. It would be great if we can find new surprises together here.

Video clipBack to the list