{"id":"arxiv/2603.08283","url":"https://arxiv.org/abs/2603.08283","tags":["cs.LG","eess.SY","math.OC"],"score":0,"title":"Learning efficient representations of complex constraints for scalable optimization","domain":"cs.LG","source":"arxiv","authors":["Yilin Wen","Yi Guo","Bo Zhao","Wei Qi","Zechun Hu","Colin Jones","Jian Sun"],"reviews":0,"subject":"cs.LG","summary":"Complex constraints often make real-world optimization computationally prohibitive at the scale and speed required for operational decision-making. Here we introduce PolyFormer, a PIML framework that learns compact polytopic representations of the geometry induced by complex constraints. PolyFormer captures constraint-induced geometry and transforms it into efficient polytopic reformulations, reducing the complexity of downstream optimization and enabling the use of off-the-shelf solvers. Neural parameterizations further enable rapid adaptation to varying operating conditions without…","category":"research","comments":0,"disputed":false,"division":"cs","mentions":0,"discussed":0,"published":"2026-03-09","reactions":{}}