archive by month
Skip to content

NLPRbot

NLPRbot is cpac by Qiyue Yin. I don’t know, but I have to suspect that it is the same version of cpac that is playing in AIIDE 2017.

It is a fork of Steamhammer. The configuration file has been incorporated into the .dll, but it still tells what is going on. Versus terran, it plays Steamhammer’s 11Gas10PoolLurker opening 90% of the time, and a couple other openings the rest (a sensible choice). I couldn’t see any difference between its lurker micro and Steamhammer’s. Versus zerg it plays a half dozen openings, the openings that Steamhammer plays most often. Versus protoss and random it plays even more similarly to Steamhammer.

It has fixed opponent-specific openings named like this, in the order listed. The opponent names are the same as in AIIDE, and some are different than names on SSCAIT.

opponentopening
UAlbertaBotOverpoolSpeedDave
Steamhammer5PoolHard
Aiur5PoolHard
Ximp2HatchMutaXimp
Xelnaga5PoolHard
Skynet5PoolHardSkynet
MegaBot5PoolHard
MicrowaveZvZ_Overpool9Gas
ZZZKBot9PoolSpeedExpo
McRave2HatchMutaMcRave

I’m not sure why Aiur and Xelnaga rated special counters. The ordinary opening mix should beat them reliably.

I didn’t dig into the dll in detail, but I do see additions that look like tracking unit types and keeping feature vectors. It looks like there are extensive changes to the zerg strategy boss; possibly a learning algorithm has been plugged in. In one game I noticed different scouting behavior. Most of the time it plays like Steamhammer.

As far as the configuration itself goes, though, NLPRbot strikes me as a mildly obfuscated fork of Steamhammer 1.3, configured for maximum wins with minimum effort from the author. I don’t know any reason other than obfuscation to rename it from cpac to NLPRbot. Maybe the author wants it to look like an unrelated bot? Maybe somebody other than the author posted it? On the AIIDE roster, Qiyue Yin is listed as “Independent”, so it’s apparently not an institutional thing.

In any case, NLPRbot aka cpac seems successful so far, even scoring a win over Krasi0. With Steamhammer skills plus hand configuration plus strategy improvements, it should be a dangerous opponent.

Trackbacks

No Trackbacks

Comments

PurpleWave on :

It looks like NLPR is part of CASIA, which AIIDE rules require to be represented to be only one entry, which was the withdrawn CasiaBot. CASIA has been a nexus of interest in StarCraft this year, producing those bots as well as (if my Googling serves me well) KillAll and Juno (Yuanheng Zhu).

The issue of what constitutes an intellectually honest fork of an existing bot is a difficult one. It's made more complicated by the time factor involved; every major fork starts out as a small one. And one author's perspective on what constitutes a substantial change may differ from another's.

Jay Scott on :

Hmm. An interesting connection. If it is related to Casia, then why is the author “Independent”?

PurpleWave on :

Because of the AIIDE rules allowing only one submission per institution to prevent collusion. There was a bit of conversation on Facebook about it all. I'm guessing part of the resolution was to allow only one entry to officially represent CASIA.

krasi0 on :

We really have to think of a way to clearly mark and force such (light) forks to announce the parent bot and original author. From your investigation, it appears that it's yet another 99.9% SH clone but there is nothing in the description field on SSCAIT page.
Wrt the win against me, my bot typically starts off by playing safely against *NEW* opponents and NLPR chose a macro oriented opening so the logical happened. If my bot had known that it was dealing with SH... :)

McRave on :

Wonder why it runs a 2Hmuta specifically against me. Is this something that is included in SH or is this just a snippet of the list of hardcoded opponents?

Jay Scott on :

Good question. Playing 2 hatch muta versus McRave is an NLPRbot decision, and not something I would have thought of. It doesn’t seem logical to me. Presumably it’s the result of some local testing and/or preconceptions, though who knows with what version of McRave.

Jay Scott on :

The opponent-opening combinations I listed make up the complete and exact list that NLPRbot has configured for those opponents, in the version that I looked at. I suspect the bot may have a greater ability than Steamhammer to break out of an opening that is not working as planned.

Antiga / Iruian on :

2 H muta done properly is quite strong vs FFE toss builds due to the mutas arriving in your base before the first sair is out. Difficult to scout as well, I could see why it is popular.

Jay Scott on :

Steamhammer doesn’t play it properly. It tends to be attracted to the cannons instead of bypassing them. Well, they adjusted the opening. Maybe they also adjusted the way it is played.

Jay Scott on :

Sure enough, the 2 hatch muta build smashed McRave, undistracted by cannons. It was an overpool fast gas build with zero zerglings, which cleared the scout with a sunken (!), not similar to any Steamhammer opening. It did make a strangely large amount of scourge, though, and kept making more after destroying the stargate. http://www.openbw.com/replay-viewer/?rep=https%3A%2F%2Fsscaitournament.com%2FReplays%2FMCRAVE%2F220813-McRa_NLPR-PvZ.rep

Jay Scott on :

Possibly NLPR stands for National Laboratory of Pattern Recognition in Beijing.

Jay Scott on :

Got it. Qiyue Yin is listed as first author on a number of papers on the page of professor Shu Wu at the National Laboratory for Pattern Recognition, which is part of the Institute of Automation, Chinese Academy of Sciences, or CASIA for short. I am beginning to understand the names and connections. http://www.nlpr.ia.ac.cn/english/irds/People/ShuWu.html

Add Comment

E-Mail addresses will not be displayed and will only be used for E-Mail notifications.

To prevent automated Bots from commentspamming, please enter the string you see in the image below in the appropriate input box. Your comment will only be submitted if the strings match. Please ensure that your browser supports and accepts cookies, or your comment cannot be verified correctly.
CAPTCHA

Form options

Submitted comments will be subject to moderation before being displayed.