AlexFiliakov/Ergodic-Insurance-Limits

AI Summary: This issue proposes to GPU-accelerate the evaluation of the optimization objective function by batching multiple Monte Carlo simulations simultaneously. This high-leverage optimization aims to significantly speed up the business optimizer, which currently runs hundreds of simulations and up to 2.5 million Monte Carlo paths per optimization run to find optimal insurance parameters.

AI Summary: This GitHub issue proposes a significant performance enhancement by GPU-accelerating the core Monte Carlo simulation loop. This loop is identified as the primary computational bottleneck, processing 100,000+ simulation paths sequentially. The goal is to transform this into a massively parallel GPU kernel to achieve an estimated 50-200x speedup.