Bridged SBI: Correcting Biased Low-Fidelity Posteriors for Cost-Efficient High-Fidelity Inference

2026-06-08

AI Summary

High-fidelity particle simulators are computationally expensive, while low-fidelity simulations introduce posterior bias due to scale simplifications, hindering accurate calibration of robotic excavation parameters. To address this, this work proposes Bridged Simulation-Based Inference (Bridged SBI), which first uses low-fidelity simulations to rapidly identify high-density parameter regions and then learns a local residual bridging function to correct the posterior discrepancy between fidelity levels, thereby guiding high-fidelity inference. This approach is the first to explicitly model and correct posterior shift in multi-fidelity simulation-based inference, avoiding the under-coverage issues caused by directly relying on biased low-fidelity posteriors. Experiments demonstrate that Bridged SBI significantly outperforms both pure high-fidelity SBI and naive multi-fidelity baselines, achieving more accurate and reliable posterior estimates with fewer high-fidelity simulations in both simulation-to-simulation and real-soil-to-simulation calibration tasks.

Abstract

Accurate calibration of particle-based simulators is crucial for robotic earthwork simulation, but analytical calibration is challenging due to this task's highly nonlinear particle dynamics and the black-box nature of conventional simulators. Although simulation-based inference (SBI) can estimate posterior distributions over simulation parameters solely from forward simulations, applying SBI directly to high-fidelity (HF) particle simulators is often computationally prohibitive. Low-fidelity (LF) simulators with coarser particles can reduce this cost, but changes in particle size and particle count shift the parameter values needed to reproduce the same observation, producing biased LF posteriors. We propose Bridged SBI, which leverages a biased but informative LF posterior to guide HF inference. This method first uses inexpensive LF simulations to identify a coarse high-density parameter region, and then it learns a local residual bridge to transport LF posterior samples toward HF-consistent regions by correcting the LF--HF discrepancy. We analyze how sequential multi-fidelity SBI (Naive-MF) can suffer from LF-induced posterior miscoverage when it directly relies on the LF posterior without discrepancy correction. We then show that Bridged SBI is designed to alleviate this issue by explicitly modeling the LF--HF discrepancy through residual correction. Experiments on both sim-to-sim particle-parameter calibration and real-to-sim calibration with real soil observation show that Bridged SBI produces more accurate and reliable HF posteriors than HF-only SBI or the Naive-MF baseline, especially under limited HF simulation costs.

Problem

• simulation-based inference
• high-fidelity simulation
• low-fidelity simulation
• posterior bias
• parameter calibration

Innovation

• simulation-based inference
• multi-fidelity modeling
• posterior correction
• residual bridging
• particle simulation