Research starts with PhD students

Chapter 49: Important leap forward research!

“A completely new algorithm concept? Is it just a concept? Not a research report or a results report?”

"I still want to hear about the bias assessment system. That part is a little hard to understand."

"Didn't Zhang Shuo say that? It's all in the paper. Just try to understand it yourself. If you can't understand it, just try to understand it slowly..."

"It's good to talk about ideas. Reports are boring. If his ideas are novel, maybe they can bring some inspiration."

The experts and scholars in the front rows had no objection to Zhang Shuo's talk on algorithm concepts; instead, they were very interested.

Just like Zhang Shuo explained, the algorithm paper has been published and the content is all there. You can just slowly understand it. What's the point of talking about the content?

The conference invited Zhang Shuo to give a report to show affirmation of the research, not to require him to talk about his own paper.

The report does not specify the content.

There was a lot of discussion in the audience.

Zhang Shuo's performance was very calm.

He waited until the discussion died down before turning around, erasing the contents on the whiteboard, and then drawing a lot of dots on it.

Then, continue to connect the dots.

The lines connected one after another formed a very complex grid, which soon became unclear.

Zhang Shuo leaned over to show the content on the whiteboard, "This is a complex network."

"As the number of objects increases, the network will become more and more complex, and the complexity will increase exponentially."

“Bias assessment is just that.”

He walked back to the podium and said seriously, "The more parameters there are to analyze, the more complicated the logical relationship will be, and the complexity of the algorithm will increase exponentially."

“And every logical relationship requires careful analysis and research to be determined.”

"Therefore, if you want to analyze large-scale particle collision experiments, relying solely on deviation assessment will be a very large project."

Everyone got serious.

When Zhang Shuo talked about an algorithm concept, some people were a little indifferent because he did not say what kind of algorithm it was. But now that he talked about the specific content, it was obviously about the data simulation deviation of large-scale particle collision experiments.

Several high-energy physics institutions are now studying data simulation deviation analysis algorithms for large-scale experiments.

They encountered the same problem -

As the number of parameters increases, bias assessment becomes extremely complex, making algorithm research a major project.

But how to solve it?
Zhang Shuo stood on the podium and gave an example that was easy to understand: "Study the economic situation of a region."

"If the target area is just a small village with dozens of households, we can go deep into every household and ask about their work, their income, their expenses, and even go into detail to ask whether they have plans to increase their income in the future, etc."

"What if the target area is a city or a state? Obviously, it is impossible to do this, no matter how much manpower is available."

Many people followed the explanation and thought about it.

Zhang Shuo continued, "To study the economic situation of a city, we need to look at GDP, employment rate, per capita income, median income, etc. We also need to use factors such as the consumer price index, Keynes coefficient, and Gini coefficient as references."

"what are these?"

“These are data, numbers!”

"If you compare two cities, the comparison is about numbers and data, not about the individuals living in the cities."

“In particle collision experiments, central anchor point connections and deviation assessments can be used to analyze small experiments, just like analyzing a small village.”

"If the subject is a city, that is, a large-scale experiment or a super-large-scale experiment, it must be compared as a whole..."

Many people have a sense of sudden enlightenment!
This is true for Bao Hexing, Roberto Castello, and James Bob.

They are backward in thinking!
Their idea of ​​using large-scale experimental data simulation deviation algorithms is still based on the methods used in small-scale experimental analysis, and they have been trying to clarify the logical relationship of the multi-parameter deviation evaluation system.

That would be a huge project!

In fact, there is no need to do that. When studying large experiments, we need to use completely new methods to look at the overall data comparison, without having to go into detail to analyze and evaluate the deviations of each particle trajectory.

But how do we make an overall comparison?

They listened more carefully.

Gao Xiaohong kept holding up her phone and filming. She didn't listen carefully to the content, but just stared at Zhang Shuo who was standing on the stage and talking on the phone screen.

The figure on the screen does not seem to be a doctoral student, but a young and successful international top scholar.

Only such top scholars can speak freely at such a conference. Even if the audience are all top scholars, they will not be affected at all and will express their ideas with confidence.

"Zhang Shuo is so outstanding!" Gao Xiaohong couldn't help but sigh in her heart, "A doctoral student can actually achieve this level." Suddenly she thought of something and sighed again, "It would be better if he was a few years older."

Then his face turned red, and he shook his head vigorously, looking to the side guiltily.

Xue Hongtao didn't care about Gao Xiaohong, because he was also shocked. The only regret was that his mobile phone was recording a video, so he couldn't share this shock with his colleagues in the group right away.

Off the script!

Don’t talk about research results, talk about algorithm ideas!

Play freely and talk freely!

It was like giving a lecture to the people in the conference hall. The problem was that the people in the audience were all top scholars, and a professor like me from a prestigious domestic university was just an insignificant little character.

Is this a report at an international conference?

Xue Hongtao thought for a long time before he came up with a similar scenario: a high school class where a professor was invited to give lectures to students.

What does that feel like?
Play freely and casually!

He also gave lectures to middle school students. He was invited by the middle school. His purpose was to encourage students to study hard and strive to get into a good university.

The current situation seems to be very similar, but the problem is that the venue is an international conference!
If it were me... cough cough.

If there is no chance to even go on stage, forget it.

……

In the back row of the conference room, Cao Mengbo looked gloomy as he walked in and casually found an empty seat to sit down.

He had just given a depressing performance report.

That was supposed to be very shocking, showing images of the Institute of High Energy Physics' deviation analysis of more than a dozen small experiments.

The conference review gave it very high praise.

The problem is that now that all the colleagues have come here, the shocking effect of the report has been greatly reduced, and it has also directly affected Cao Mengbo's mood.

He didn't even want to listen to the comments and came here directly after finishing his report.

After Cao Mengbo sat down, he looked around and asked Tian Guilin, who was sitting one seat away, "How long has the report been going on?"

The person next to him immediately said in dissatisfaction, "Quiet!"

Tian Guilin pointed at the stage and reminded quietly, "Listen carefully."

Cao Mengbo suddenly became even more depressed. He could only turn his head and listen carefully, only to find that it was not the center anchor point connection algorithm that was being discussed?

On the stage.

Zhang Shuo has already mentioned the key point, "I think the best way to analyze the data simulation deviation of large-scale experiments is to make a numerical comparison."

"Comparing numbers is not like analyzing the city's economy, there is only a general number."

“We can convert experimental simulations into numbers, and then through comparative analysis, we can convert experimental data into numbers, and use simulations to supplement the missing experimental data, so as to conduct overall comparison and overall output.”

"There are a lot of benefits to doing this."

"The first is that it can simplify the algorithm. Using the central anchor point connection algorithm to study the analysis algorithm of large experiments, even if it is successful in the end, the completed algorithm will be very large, and it may even be so large that even the most advanced computers cannot complete the calculation."

Many people nodded along with the explanation.

This is the key issue.

If the algorithm is too complex, the amount of calculation will become very large. Even if a perfect algorithm is developed, it may not be put into use at all due to the unpredictable amount of calculation.

“Second, a holistic analysis can confirm the discovery.”

"The results of experimental bias analysis reveal that special phenomena in a single region can be largely attributed to external influences."

"For example, the first small-scale experimental analysis by the Institute of High Energy Physics found that there was a small area with a small number of particles. The conference concluded that this was due to the presence of air molecules in the vacuum pipe."

"But what if it is an unknown physical phenomenon? How can we determine it? Obviously, we must conduct a holistic analysis of large experiments and a holistic analysis of many experiments."

"This is the only way to increase the credibility of the results and determine what the special phenomenon is..."

When the explanation reached this point, the expressions of the data experts in the front rows had become extremely serious.

At the beginning of the report, they just thought it was interesting to hear about the "algorithm concept", but later they found that the algorithm concept might solve the problem of deviation analysis calculation in large-scale experiments.

Now, they believe that this will become an important research, and may even become a great achievement that has a leap-forward impact on high-energy physics experiments!

(End of this chapter)