用 GPT-6 Astra 讨论出在线逆线性优化最优遗憾界

Shinsaku Sakaue @shinsaku_sakaue原推发布于 2026.09.24

在 X 上打开原推

After extensive discussions with GPT-6 Astra, I've closed the regret gap in online inverse linear optimization! For utility and actions in the d-dimensional Euclidean unit ball, the optimal regret is Θ(√d). The algorithm is far from efficient, though. 🔗 https://arxiv.org/abs/2609.26978

这条是引用着另一条写的：x.com/shinsaku_sakaue/status/2100365620999245842