08
Jul
[Submitted on 4 Jul 2024] View a PDF of the paper titled NeuroSteiner: A Graph Transformer for Wirelength Estimation, by Sahil Manchanda and 4 other authors View PDF HTML (experimental) Abstract:A core objective of physical design is to minimize wirelength (WL) when placing chip components on a canvas. Computing the minimal WL of a placement requires finding rectilinear Steiner minimum trees (RSMTs), an NP-hard problem. We propose NeuroSteiner, a neural model that distills GeoSteiner, an optimal RSMT solver, to navigate the cost--accuracy frontier of WL estimation. NeuroSteiner is trained on synthesized nets labeled by GeoSteiner, alleviating the need to train…