Amortized Neural Optimization for Pre-Layout Signal Integrity Design Space Exploration using Differentiable Surrogates
2026-06-05 • Computational Engineering, Finance, and Science
Computational Engineering, Finance, and ScienceMachine Learning
AI summaryⓘ
The authors address the challenge of optimizing high-speed signal designs, which usually takes a lot of computing time due to repeated simulations and searches. They propose using a method called amortized neural optimization (ANO), which trains a neural network to predict the best design settings all at once, without repeated searching. This approach drastically speeds up optimization, making a task that normally takes days finish in milliseconds, while only sacrificing a small amount of accuracy. Their tests on different signal integrity problems show that ANO can make pre-layout design exploration much faster and more practical.
signal integritydesign space explorationneural networksamortized optimizationelectronic design automationblack-box optimizationdifferentiable surrogate modelsDDR5 DFESerDes co-equalizationeye diagram
Authors
Julian Withöft, Werner John, Emre Ecik, Ralf Brüning, Jürgen Götze
Abstract
Pre-layout design space exploration (DSE) for high-speed signal integrity (SI) analysis is often limited by the computational cost of simulations and iterative optimization algorithms within modern electronic design automation (EDA) workflows. While machine learning surrogate models accelerate the simulation step, optimizing designs still requires utilizing iterative black-box search methods. This iterative nature scales poorly, making multi-corner sweeps computationally expensive. As a solution, this paper proposes amortized neural optimization (ANO) for pre-layout SI design. ANO entirely eliminates iterative black-box inference by utilizing fully differentiable neural network surrogate models. ANO extracts analytical gradients from the surrogate to train a global optimization policy. Instead of solving the optimization problem repeatedly at inference, the optimization process is learned offline and therefore amortized. Once the ANO policy is trained, it maps different channel contexts directly to near-optimal design parameters in a single deterministic forward pass. The efficiency and accuracy of the ANO framework are demonstrated based on three complex SI design scenarios, including DDR5 decision feedback equalization (DFE), 9-dimensional SerDes Tx/Rx co-equalization, and DDR3 DQS differential pair routing to optimize eye diagram metrics under intra-pair skew constraints. By trading roughly 10% in optimality compared to instance-specific black-box algorithms, it realizes speedups of three to four orders of magnitude. For a large-scale 320,000-instance multi-corner SerDes sweep optimization, ANO collapses what would have taken days of computation using iterative search algorithms into a single batched forward pass that completes in milliseconds. This transforms computationally expensive SI optimization into real-time and interactive pre-layout DSE.