Spectrally accurate reverse-mode differentiable bounce-averaging operator and its applications

We present a spectrally accurate bounce-averaging operator implemented as a part of the automatically differentiable DESC stellarator optimization suite. Using this operator, we calculate the proxy for neoclassical transport coefficient $\epsilon_{\mathrm{eff}}^{3/2}$ in the $1/\nu$ regime and benchmark it against the NEO code. Ultimately, by employing this differentiable approximation, for the first time, we directly optimize a finite-$\beta$ stellarator to enhance neoclassical transport using reverse-mode differentiation. This ensures that the computational cost of determining the gradients does not depend on the number of input parameters.