Towards Robust Learning to Optimize with Theoretical Guarantees | Read Paper on Bytez