TNTHead wrote:Hoffman900 wrote: ↑01 Feb 2026, 21:59
FW17 wrote: ↑01 Feb 2026, 18:43
AI will generate the most optimum generate and deploy map.
I work in engineering and while AI might be able to crunch a lot of situations relatively quickly, it will not generate the most optimum anything without human intervention.
It certaintly doesn’t work like that.
Indeed but after a few races with the data per cornertypes and connecting straights you can train a model or optimisation algorithm that can optimise upcoming tracks. May be not as refined as the real world data, but you probably can come close.
Not really. Data makes it easier but we are talking about being 1% off puts you well behind.
These are structured problems, the best solution is given by optimizing a formula that gives correlation between the variables in play and looking to maximize the integral of the variable you care about.
Data can help create the formula that correlates your variables, but you still need to do the work and further iterate. ML is a tool like others that needs multiple iterations to work, implemented by people that don’t know what they are doing is a disaster.
