ScalarR/Some Ideas

This article is a stub. You can help RoboWiki by expanding it.

I’m not going into full details for now, but rather some crucial ideas that drives ScalarR and have been proved to work well. Instead of encouraging people to follow the ScalarR way, I believe that sharing initial motivations could inspire even more innovations, and push the bar of top bots even further.

Targeting

Targeting is essentially modeling the physical thing in some other space, e.g. GuessFactor space. Everything happened in physical space have some projection in “targeting space”, and vice versa. By designing features, Kernel Density Estimation, lateral direction, etc., you are essentially building some relationships between the two spaces. So it’s best that each point in physical space that matters for targeting have some unique representation in targeting space, and points in targeting space can be mapped to physical space precisely, or exactly. In simple words, everything blends in harmony. This is the most important thing in targeting, even more importantly than things like “KNN weights” etc. By keeping this in mind, you easily filter out targeting models that are essentially bad, leaving only the most possible ones.

Surfing

Surfing is another story, a complete different story. But if I was asked to give Wave Surfing another name, I would simply choose Minimum Risk Movement, redefined. Back to the simple story where agent (robot) was facing an environment (mainly walls and other bots), which action (mainly movement) will you choose to maximize objective, e.g. survival? This is not a simple question, but many non-simple questions receive some simple solution — why not simply simulate what's going next, and let the simulation result decide? Then many techniques are used, e.g. Precise Interaction, Bullet Shadow/Correct, everything is making the simulation closer to what's truely happening, by making it more precise, and more exact. But there's one thing left, something is unknown either in melee or in 1v1, and that thing is what separates a top bot from the others — danger estimation.

Instead of inventing and experimenting some novel approach like in targeting, the best way that worked is to test against real opponents, say the entire rumble, since danger estimation is essentially fitting the targeting of existing robots in the rumble. Inventing something fancy doesn't help much, but waching a lot of battles, reading a lot of targeting code does work well. And once most problem bots are solved, you got some fairly good result, done.

Energy Management

Energy Management is not simply Power Selection. By doing energy management, you are solving the same problem as in surfing — what's the best action (now bullet power) that maximizes objective, e.g. survival. This involves precise simulation sometimes but simulations are much harder here given the nature of random process. And most observations are biased, so do most simulations. Some very simple strategy possibly in one line can give you astonishing result, but coming up with it may require days of watching battles, and even more days of failing attempts. And past tuning of energy management gets outdated soon after mutating surf & targeting. So this is generally the last thing I do. And keeping it as simple as possible until the very last time does save you a lot of time.