2024 ICPC Shenyang Travelogue
Summary: A humiliating defeat.
Day -1
On Thursday, I even specifically skipped the evening general education class, staying in the training team classroom from the afternoon all the way until night, even though my mind was already in a chaotic mess by then.
I really, really, really, really, really, really, really, really, really, really wanted to win a medal. I really wanted to win a bronze medal. A glittering bronze medal. But what is the use?
That day, the airline canceled the return flight for Sunday night, so hyl and I rescheduled. chy, however, due to an online exam or something similar on the weekend, chose to come back on Monday instead.
Then I discovered that five notebooks of templates and one copy of oi-wiki weighed slightly more than 20 jin. That day, I even asked the team if they preferred 0-indexing or 1-indexing, only to realize that the specific indexing depends on the specific problem.
Day 0
We set off for hgh on Friday morning and found that the check-in counters were different from what I had seen in videos; they had all become self-service. And later, when returning from Shenyang, I found that arriving at the airport just half an hour in advance is sufficient, and security checks take no more than ten minutes.
That day, I even stuffed a sweater into my bag, only to find it completely useless. After landing, I discovered that a down jacket was quite enough. It wasn’t cold at all.
At seven o’clock that evening, the three of us started a vp of last year’s Shenyang contest in the hotel. The bad news was that we discovered we had done the individual contest problems before… The good news was that we got bronze with 3+1 problems.
Then we were roasted by the seniors in the group chat for doing a vp while out on a trip. Looking back now, I guess the pressure was just too high at the time. If we had relaxed a bit, maybe we could have performed better. I have to say, the seniors really have foresight.
Day 1
chy and I slept until 10 o’clock. Originally, we wanted to use the morning to go out for a stroll or practice some problems, but the morning was just gone.
In the afternoon, check-in, warm-up contest.
Looking at the first problem of the warm-up contest, such a lengthy problem statement, definitely didn’t need to be read. Then flipped to the back, it was an interactive check-in problem, just ask 100 times. However, I did remember 1923, a year that is perhaps very important to Northeastern University.
Then, for the second problem, I misread the problem like a war criminal, thinking it was that computational geometry problem from Shenyang '23 that no one in the entire venue solved. Later I realized the original problem was expectation, and this problem was the expectation of the square; x and y could be separated based on integration properties. I didn’t dare to think about it on the field.
The third problem came from Zhihu, a dp, didn’t understand it. Only Jiang Li’s team solved it.
The fourth problem was a Nonogram. Our team was very lucky to have brought two pencils (hahaha); I can’t imagine how to do this thing without pencils. hyl and chy filled it out in the last five minutes. Afterward, hyl complained to me that he could practically memorize the numbers.
Anyway, we got bronze in the warm-up contest relying on two problems, just like the warm-up bronze in Hangzhou last time. Just like the Iron in the main contest last time.
Day 2
The seats were very cramped.
Sitting behind us was the Laisheng team from Zhejiang University. Then we couldn’t hear anything.
We finished the check-in problem in just 8 minutes, 1 minute and 2 minutes faster than Team 2 and Team 6 respectively, but it was of no damn use. Team 2 took the First Blood for Problem B.
After that, we looked at Problem B for quite a long time. It seemed the construction was correct, but the calculation was wrong, leading us to believe our construction was wrong. By the way, regarding Problem B, I saw other teams saying it was a conclusion similar to CRT. I’m ashamed; I had recently talked about CRT-related stuff with chy, but neither of us realized this.
Problem E. After 3 hours, I wrote an state compression DP. It TLE’d. I didn’t think of the suffix min optimization. Oh goddddddddddddddddddddddddddddddddddddddddd, missed it by just that little bit. I don’t know if the simulation chy wrote was correct, but I felt it was quite sketchy. At that time, I only thought about recording the transition of a certain state to the final state, but didn’t think of recording the transitions of all states in one go. As long as special handling is done during elimination, it could be optimized to to answer.
Problem D, my two teammates quickly thought of the inversion pair conclusion, but later kept thinking about how to maintain the specific number of inversion pairs, and even thought about how to specifically maintain the shift operations. This is probably impossible; we simply didn’t think that considering parity would be enough. Still too little experience and too nervous.
Problem B, constructed half of it in the last minute, estimated to be beyond redemption.
In the late stage, my whole person was overheating. chy kept arguing with me about the approach for Problem E, it was very chaotic. Also, when he was writing the dfs, I should have been watching; it shouldn’t have come to him writing dozens of lines for operations that could be compressed into a few lines of XOR. Mainly, I was very irritable at the time, very irritable, very irritable, finding it hard to accept the fact that we were going to get Iron again. Instead, in the last few minutes when I knew we were truly doomed, I calmed down a bit and made some progress on Problem B. For the first few hours, I was constantly overheating, under massive pressure. I discussed my state compression approach for Problem E with hyl, but didn’t reach any conclusion. I also kept feeling that chy’s approach was too rough; we just didn’t see eye to eye. If only T=1. T=10000 was too much of a scam. There was even a calculation error of 2^16=256 at one point. The shitty computer froze at any moment; downloading sample data could cause several windows to lag out. Next time, I need to try eating something during the contest to regulate my emotions. Next time is very far away.
The key point was that I couldn’t falsify chy’s approach, only relying on a mysterious and abstruse feeling that it wasn’t quite appropriate. On the field, I also never thought of how to optimize it, even thinking about how to optimize the exponent 16 to 9, rather than the actual intended solution of 0. At that time, I did actually think that the intended solution was to lift the entire dp process outside the query, but I simply didn’t think of how to handle the queries in that situation, because I simply didn’t think that all query states should be handled at once instead of opening a separate state for each query. Different states can clearly contain each other.
Just solving the second problem before the three-hour mark would have secured a medal. And those three problems all had a chance, all were only half done, none were hard, none were hard, none were hard.
But what is the use? Even if a bronze medal is useless, isn’t having not even a bronze medal even more of a waste?
After finishing, hyl and I rushed to catch the plane at light speed, only to find we were an hour early.
Anyway, we left this city that broke our hearts.
Conclusion
The revolution hath failed utterly; comrades must still strive on.
Anyway, next up is just playing cf; there are no main contests left to play. Yesterday I played a div1+2 and only solved 2 problems. I guess the pressure was too great; I felt my brain just wouldn’t turn.