Spark

Chapter 169: The Inception of Convolutional Neural Networks

Looking at the radiant Wang Jiancheng, Chen Nian felt that this kid was really not simple.

If he thought he was chasing his dreams when he proposed a series of future weapon concepts before, then after chatting with him today, I am afraid that he will have to change his wording.

It's not chasing dreams, it's making dreams.

Difficult, but not impossible.

How are you going to do it? Drone combat is a complex system, especially when it comes to coordinated operations.

You don't have a little understanding of the data link system, and your knowledge system is not complete enough. It's too difficult to start a loyal wingman plan, isn't it?

Hearing Chen Nian's words, Wang Jiancheng nodded and replied:

Senior, I also know it's difficult, so I didn't intend to make the project too complicated from the beginning.

What I want to do is a remote control system that responds in a timely manner.

In general, it is necessary to integrate the control of the drone into the existing fighter avionics system, and realize the actions of the drone to follow, accompany, attack, and cover through simple control.

When the technology is more developed in the future, use this as a foundation to implant various tactical templates to achieve one-click control.

For example, when our satellite and data link systems evolve to be sensitive and powerful enough, we can even equip a J-22 with several drones, including electronic jammers, small early-warning aircraft, and ground attack aircraft. machine, and shield machine.”

. Awesome.

Chen Nian sighed sincerely.

What makes this set of ideas awesome is not his idea of ​​drone tactics, but a sentence:

system integration.

He probably thought that what he was doing was very simple, but in fact, he was really digging the foundation!

If this set of integrated control system can run smoothly and carry out open iterative development, maybe the first generation of loyal wingman of the Republic will really be born in his hands.

Have any ideas?

Chen Nian continued to ask.

Yes!

Wang Jiancheng smiled, pushed away the already eaten dinner plate, and answered:

My plan is this. This system is simple to say, but it is actually quite huge in practice.

So, I don't have to do it all at once, so I just need to make a simple imitation and following part first.

The implementation method is also very simple, directly through data synchronization, but the current problem is that there are differences in performance, speed, and flight envelope between the drone and the host. It is necessary to use algorithms to bridge this gap. It's easy.

You still know how to program?

Chen Nian asked in surprise.

Understood, my dad bought me a computer when I was very young. At that time, I sold the orchard to buy it. There is no other use, so I taught myself programming.

what language?

C#, python.

Awesome.

In just a few minutes, Wang Jiancheng made Chen Nian sigh several times.

However, thinking about it carefully, it was his father who was even more awesome.

Wang Jiancheng's family conditions are not good. There is an older sister in the family, and the financial pressure is not small. Under such circumstances, he dared to make such a big gamble on educational resources, which shows that he is definitely not a short-sighted person.

This can be regarded as a tiger father without dog son to some extent.

Then what algorithm are you going to use to solve it?

Uh, what I'm currently thinking about is machine learning, so that the wingman has a certain decision-making ability as a supplement to imitation. The main use is ensemble learning, specifically the random forest algorithm, but I feel that this algorithm is still a bit weak.

There are no more good ideas for now, let's take a step and see. I think even if this is the case, it is enough to enter the special class.

That is certainly enough.

Even if it's not enough, I'll let Chen Guo arrange it for you.

Chen Nian thought to himself.

However, the random forest algorithm seems to be a little behind in this era.

In my memory, by next year, that is, in 2006, deep learning algorithms will appear.

Instead of letting him engage in random forests, why not go directly to deep learning?

This algorithm is not only more excellent in machine learning, it can even be said that the emergence of deep learning algorithms has triggered a frenzy of AI explosive growth.

If Wang Jiancheng is really allowed to do it, the future is immeasurable.

Even if it's just a start and a foundation for future generations, it's still a good attempt.

Thinking of this, Chen Nian said:

I have an idea, you can try it.

Now there should be some machine learning training models about layer-by-layer initialization appearing on the paper library. Based on this model, there are more potentials to dig deeper.

With your ability, it shouldn't be too difficult to use ensemble learning at this age and continue to explore in depth.

You can first look at the most advanced experience in the world, and try to use it in your own practice.

Among them, you can focus on the SVM method. If you have difficulties, you can ask a teacher majoring in statistics for help.

Having said that, Chen Nian paused.

He suddenly thought that at this point in time, it should be difficult for the other party to find more useful materials.

However, if you let yourself parse this part of the content and give it to him.

Wipe, 6.9 source points are consumed.

Then pull it down.

So, he recalled carefully for a moment, and continued:

There are a few keywords that should be useful in your research.

Kernel function expansion, linear regression function, recursion forget it, I will draw a rough schematic diagram for you to see.

Chen Nian picked up his chopsticks, dipped them in the soup and began to write and draw on the table.

What he drew was the basic structure diagram of the most commonly used convolutional neural network for deep training of game AI.

In fact, the principle of the entire convolutional neural network is not complicated. The most basic part, even ordinary people who have no programming foundation at all, can understand it in half a day.

The real difficulty is how to implement it with complex programs.

This is not what Chen Nian is good at. In the last life, he just learned a little out of interest.

Here, this is the filtering process diagram of the convolutional neural network. This is the C1 layer, which is used to generate a feature map, and then the features are summed and weighted.

The soup gradually formed a criss-cross picture on the table. Wang Jiancheng had been listening to it wholeheartedly from the very beginning, so even if Chen Nian only said it once, he had already memorized the whole content in his heart.

When Chen Nian put down his chopsticks, the way he looked at Chen Nian had completely changed.

At first, I thought that this was just an ordinary, enthusiastic senior with a little achievement.

But now it seems.

What kind of monster is this!

He said that he didn't understand programming, but the backhand was an algorithm that was obviously better than the random forest.

Although this algorithm is indeed not up to the level that can be used immediately, his thinking is definitely top-notch!

You also said you don't know martial arts! ?

Wang Jiancheng carefully wiped off the excess part of the picture with a paper towel, not knowing what to say for a while.

The students passing by looked at this scene curiously, with funny expressions on their faces.

He probably felt that the two of them were crazy about their studies, and their reasoning was pushed to the dining table.

But what he didn't know was that the picture on the dining table might become a well-known story after many years

After bidding farewell to Wang Jiancheng, Chen Nian strolled back to his residence by himself.

In fact, revealing the information of the convolutional neural network to him is not a too advanced technology disclosure, because as early as 2002, the basic principles of the convolutional neural network were born.

Until the outbreak period in 2006, most of the work of scholars in this process was actually supplementing and perfecting the basic principles.

What I did just now was to convey what I already had to Wang Jiancheng. As for how to actually apply it, he still needs to explore.

Really speaking, such an action is actually equivalent to planting a seed.

According to Wang Jiancheng's talent, even if the seed is not planted by himself, it will be planted by other people or other things during a certain study or seminar.

Therefore, in general, the risk is very small, but the return in return is very large.

but

Why is the system not responding!

Wang Jiancheng, Wang Jiancheng, can you do it?

Others always break through as soon as I mention it, you still have to digest it, right?

I hope you don't let me down, if you can give me another surprise, that would be even better

Back at the residence, Chen Nian slowly opened his textbook, preparing to continue learning his advanced engineering materials.

The progress of this course has reached nearly 70%. It is estimated that in another month, the entire course can be completed. At that time, you can start to try to switch to the rhythm of scientific research + learning.

With the support of the system, his research on new materials will definitely be much easier than others, and the source of knowledge increments in the process can also make up for his own consumption.

Virtuous circle!

Chen Nian nodded in satisfaction, turned around and devoted himself to studying.

And at this moment, on the other side of the sea, in the office of the May Society.

Paine sat decadently on the sofa, his eyes were full of frustration, as if he had lost all light.

He couldn't figure out why the plan this time failed again.

The two generals I personally sent out used the most concealed and safest strategy, but in the end, even they themselves fell into it.

Bond and Sean are all arrested!

How keen is the perception to be able to locate the instigator after something happened and before one's own side is aware of the danger, and even arrange everything to arrest people?

In all fairness, even my own side cannot achieve such efficiency and accuracy.

On the other side of him, Houston had the same expression.

Looking at the report in his hand, he sighed and said:

The failure of this operation has had a huge impact on us.

The lists we're forced to exchange will deal a major blow to our intelligence community over there.

According to the current situation, it will take us at least four or five years to restore the efficiency of intelligence operations before.

Also, it's not so easy to deploy new personnel there now, especially for the deployment of the other party's project, which has basically become impossible.

Hearing this, Pan En's face became more and more gloomy, but this was far from over.

What's more serious is that after this failure and the impact of the previous incidents, Congress's trust in us has been greatly reduced. If there is no result, it is very likely that our budget will be seriously affected next year. reduce.

After finishing speaking, Houston hesitantly put down the document in his hand.

In fact, he had even worse news, but in front of Pan En, he couldn't say it.

The congress's resentment was not only towards the May Society, but more towards Paine himself.

His situation is very dangerous, and some people even asked whether he might be a spy who has been lurking for a long time like Jin Wushao.

If it wasn't a spy, how could it be possible to make so many stupid mistakes in a row?

Thankfully, such views are not supported by the majority, as they are well aware:

A real spy wouldn't be so stupid

The two were silent for a moment, and Paine asked:

What we're failing at is the final solution, right? What about other conventional actions?

...it didn't work. We did create a firestorm, but in reality, their decision-makers didn't seem to care.

The project is still advancing, and the returnees have basically arrived in Chengdu. Even in the near future, their material consumption has shown signs of accelerating.

At this point, we really underestimated them.

Paine let out a long sigh and continued to ask:

What about the other projects? Gas turbines? High-speed rail?

We didn't have the opportunity to intervene in the high-speed rail part. In addition, they have already mastered their own core technology. The matter is basically a foregone conclusion.

Alstom and Kawasaki Heavy Industries cooperate with their CRRC and CNR respectively, and Bombardier cooperates with CSR.

They were unstoppable.

Houston carefully observed Paine's expression, and then said:

As for the gas turbine project. Our layout at the Motor Sich plant is going well.

The model is already in hand, but

but what?

Paine asked, frowning.

However, after this period of verification, the model fed back by NAS is likely to be wrong.

Maybe it's a trap in itself.

Hearing this, Paine stood up abruptly, staring at Houston, and asked word by word:

Is there any other bad news?

Houston shook his head quickly and replied:

temporarily unavailable.

Paine heaved a sigh of relief, secretly rejoicing in his heart:

Fortunately, although the situation is serious, it is not out of control.

He looked at Houston and said in a low voice:

The problem with the model is not too big. It's not necessarily a trap. It may be that they themselves have flaws in their design. With a little time, we can fix it.

In addition, we can still recover the equity of the Motor Sich plant, so as to avoid greater losses.

I'm afraid not.

Houston said in a bitter tone.

Why?!

They launched the poison pill plan, and now the operating conditions of the entire Motor Sich plant have dropped to a historic low.

It is already heavily in debt, and most of the equity has fallen into the hands of Dongqi Group. Even if we take it back now, what we get is nothing more than debt.

Paine was stunned and stunned. At this moment, a strange thought even occurred to him:

In order to obtain the technology and equipment there, and to fight for a chance to break through the barrier, they are actually willing to do this to this extent

With such courage, losing to them seems to be a matter of course?

He felt his legs go weak.

Intense dizziness rushed to his brain, and after a while, he passed out completely in this dark, hidden but high office

Tap the screen to use advanced tools Tip: You can use left and right keyboard keys to browse between chapters.

You'll Also Like