I Became a Genius Engineer - I Became a Genius Engineer chapter 95

Artificial Intelligence Developer Cheolwoo Lee (19)

I thought Minseok was going.

‘It was easy to say. Treat the shard network as a single neuron and connect it through synapses.’

In fact, from the time of making automatic driving artificial intelligence and interpretation and translation artificial intelligence, how to do strong artificial intelligence. In other words, I have been thinking a lot about whether I can learn and have the same way of thinking as humans.

As a result, it was concluded that the NF network should become one giant brain and connect the shard networks under it with neurons and the aggregate of shards through synapses.

‘Even that is just a theory.’

It is difficult to realize it right away, but I couldn’t erase the question that the bigger problem could be created by realizing it.

Surely there have been times like this.

‘ITE-X engine, this is what I felt when I was researching it. It’s depressing.’

Before making the 5-stroke ITE-X engine, the 2-stroke ITE engine had unfounded confidence, but it failed spectacularly.

And the 5-stroke engine, ITE-X, which was created after standing up.

I felt the same way when I made it. The ITE-X engine, which must be made from scratch… … . It was stifling.

How on earth did they come up with it back then?

Thinking about it now, it was amazing.

The ITE-X engine and the current strong artificial intelligence were also vague, but the lack of confidence was the same.

I was able to create the ITE-X engine thanks to a lot of hard work and, in a way, luck… … . The current strong artificial intelligence is well.

‘Ah, why did you say that.’

Now that I think about it, I feel a bit regretful. It is said that it is making strong artificial intelligence, and it is receiving strong support from Google, but what if development fails?

What position will Google take?

Not only that, but if that degree of technological super-gap is not manifested, China will continue to act again.

‘No, this is not the time to think negative thoughts like this. let’s do something I have a plan.’

started working right away.

a few days to do that.

‘I guess I’ll have to invite an expert.’

admitted to himself I’ve solved almost everything by myself until now, but this time it’s a little different. I thought it would be difficult to create without help.

I immediately called Minseok.

-Uh, Cheolwoo~ Is your research going well?

Minseok said in a pleasant voice.

“I have something to ask for a little help with.”

-help? What is it? Just tell me what you need and I’ll get it for you no matter what.

“I need advice from an artificial intelligence expert. Professor Jeffrey Hinton, can you invite him?”

-… … ? Need expert help?

Did I, who had never received help from an expert, be puzzled when I tried to invite an expert? Minseok asked.

“Uh, I guess. Rather than acting alone, I need advice.”

– How can I say that I am proud? Until now, I tried to handle everything by myself, but when it was difficult, I knew I would get help from others.

“Oh, okay. Okay, no? Just say that.”

-hmm… … . It might be a bit difficult though.

“Is it too difficult?”

-Anyway, we signed a cooperation with Google after loudly saying that we could create strong artificial intelligence. But on the contrary, try to ask for help.

“Huh, right?”

– But I’ll find a way. If need be, I’ll try to save it somehow. Because I understand very well how difficult strong artificial intelligence is.

“I got it, thank you.”

Then I called a few more people.

A competent developer was needed.

* * *

Seo Jeong-ho, who became Hyun-sung’s vice president of technology, has been extremely busy lately.

Previously, I had to focus only on research, but now I had to worry about everything in making a car, including research, production, and verification.

And he visited the plant of Japan’s recently acquired Toyo Motors to check out some problems.

“Vice President Seo. Are you here?”

“Yes. Hello. I heard that labor costs are high these days?”

The problem is the labor cost.

Labor costs were too high not only for factories acquired by Japan’s Toyo Motors, but also for factories in Hyeonseong.

So, I recently planned to establish a factory in China or Vietnam, but… … .

‘If you build a factory in a place like that, you know when you’ll get hit in the back.’

The risk was too great. In particular, China had already had dozens of precedents, so it was not considered, and even other countries were dangerous.

To that end, building factories in advanced countries such as Japan, the United States, and Europe had no merit in terms of labor costs.

So, to solve the problem, Vice President Seo was visiting factories all over the world.

“Yes. That’s right. Making the X-380 requires more manpower than expected.”

It wasn’t just the X-380. All cars equipped with ITE-X engines require more manpower than conventional cars.

Existing production facilities could not produce the ITE-X engine, nor did the X-380.

Even if a dedicated production facility was created, the labor cost would be higher than before.

The point was that there was a limit to production even though a huge number of orders were coming in.

Of course, as a temporary countermeasure, various methods were being implemented, such as raising the price of vehicles and hiring more people by raising labor costs, but it was only a temporary countermeasure.

“hmm…….”

While touring the factory, Vice President Seo checked the structure that required a large number of manpower and fell into thought.

‘How can I solve this? Should I seek advice from Mr. Lee Chul-woo?’

I thought about that, but I thought that it was not just about visiting teacher Lee Cheol-woo for this problem.

‘Well, artificial intelligence is being developed right now. Wouldn’t it be possible to use that technology?’

He even considered using cutting-edge technology, artificial intelligence.

‘Obviously, Researcher Jung said that it couldn’t be used for things like this.’

Medicine artificial intelligence has its limits.

‘What did you say? Did you say that it would be possible if strong artificial intelligence was created?’

Vice President Seo pondered how to solve this problem.

* * *

Again, a few weeks later, several people were dispatched to my laboratory.

first.

“Hello, this is Jeffrey Hinton. Nice to meet you as a fellow AI researcher.”

Professor Jeffrey Hinton. He was the founder of the deep learning paper.

How the hell did Minseok seduce him?

“Hello. I’m Lee Cheol-woo.”

“Translation artificial intelligence, it’s my first time using it, and I really like it. It’s great.”

“Ha ha, thank you.”

In one ear of mine and in the other of Professor Geoffrey Hinton, earphones were fitted, and the translated voice came through.

“It’s really amazing. I made a theory, but how can you design artificial intelligence so completely and exactly as it is in theory? Inventor Lee Cheol-woo is really amazing.”

From noble mtl dot com

Is this time an inventor… … .

“thank you.”

It is certainly not an easy task to implement something as it is in theory. I took his praise for what it was.

“Hello. Nice to meet you. I’m Vitalik Bute.”

And Vitalik Bute, a former Ethereum developer who recently joined the NF Foundation as a developer. I told the NF Foundation that I was dispatched for a while.

I am the only person who is familiar with the blockchain network.

He, too, was wearing an earphone in one ear and was using translation artificial intelligence.

“hahahaha, nice to meet you. I’m Lee Cheol-woo.”

“NF Network is really great. It’s the ideal blockchain network I’ve been thinking of.”

“It seems odd.”

“That’s right. It’s fair and transparent, and decentralization will be the trend in the future.”

“Yes, it’s already going on.”

“Thank you for inviting me to the AI development team this time. I will do my best.”

“Yes. Please take care of me too.”

Sometimes I thought it was sad. Without me, Ethereum could have taken the second place after Bitcoin.

Does home mean nothing?

“Hello. My name is Suhyuk Nam, who majored in robotics and ergonomics. I am currently researching the humanoid arc.”

Lastly, Dr. Soo-Hyeok Nam.

Dr. Soo-Hyuk Nam, a graduate of KAIST, is working as the head of the humanoid laboratory.

It seemed to help in many ways.

Artificial intelligence expert, Professor Jeffrey Hinton. Blockchain developer Vitalik Bute. Dr. Suhyuk Nam majored in humanoid and ergonomics.

The best members gathered.

It was judged that strong artificial intelligence would not be complete without this level of experts gathering.

“Thank you all for coming. Everyone said hello, let’s go see the artificial intelligence I’m designing right away.”

After each answer, they entered the small seminar room.

And I started presenting right away.

“What we are currently planning is artificial intelligence that can think and learn like a human. The first design model was named NCI model that imitates neurons and synapses. So, in the end, a giant artificial intelligence resembling the human brain will be completed. However, there are several problems, first of all, there are transmission delay and transmission speed problems, and solving them is the task of this project. It is. Let’s go into a little more detail…”

A fairly lengthy presentation ensued, and the three experts I invited listened intently.

“……. This was the outline of the NCI model.”

pair- pair- pair-

At the end of my presentation, the three experts stood up and applauded.

“Wow, that’s great. How did you come up with that idea? By substituting neurons and synapses into the sharding network of the NF network, I was able to think perfectly. In particular, the x2xx algorithm used to neuronize the sharding network is so amazing. And After that, the theory of critical thinking that connects synapses is a Nobel Prize.”

said Dr. Jeffrey Hinton.

“I especially focused on the NF network. I was surprised that such an implementation is possible in the current Flyer language. It is impossible to develop with existing programming languages. Because of the NF network, I can create the x2xx algorithm.”

This time, Vitalik Bute said.

“Well, I have a question. The theory of critical thinking that Professor Jeffrey Hinton talked about earlier. Can you elaborate on that?”

It was a theory that became the core of strong artificial intelligence.

“Yes. Critical Thinking Theory, current weak artificial intelligence, is being achieved through Professor Jeffrey Hinton’s deep learning, or deep learning. It is an algorithm that learns by finding commonalities in a huge amount of data. However, strong artificial intelligence cannot make it that way. I decided there was no such thing, so a new theory was needed.”

“That’s right. It’s a theory I came up with, but it’s not a theory that can create strong artificial intelligence.”

Professor Jeffrey Hinton helped.

“So a new theory was needed, and that is Critical Thought Theory. To explain it a little easier, how can a baby learn to speak?”

“Yes, it must be through learning.”

“So here’s the question. How long will it take for a baby to be able to speak?”

“It takes roughly a year and a half.”

“Then, how much information is the child listening to in terms of quantity?”

“Um… that’s fine.”

“I have published statistics before. There are errors, but a child listens to about 200,000 sentences while living with a family for 1 year and 6 months. And that data is much smaller than the development of artificial intelligence.”

In fact, about 10 million photographic data were required just to distinguish plants by artificial intelligence, especially recently developed automatic harvesting robots.

Interpretation and translation artificial intelligence also required as much data as that.

By comparison, 200,000 was a small number.

“In order to create strong artificial intelligence, it was first necessary to increase the learning efficiency. Critical Thinking Theory was born as a result.”