# ASBI: Leveraging Informative Real-World Data for Active Black-Box Simulator Tuning

arXiv: https://arxiv.org/abs/2510.15331 · HTML: https://mlsanity.com/paper/2510.15331/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 57 references that are in the map.
- First posted on arXiv 17 Oct 2025.

## Builds on (most cited first, 18 of 18)

- [The frontier of simulation-based inference](https://mlsanity.com/paper/1911.01429/) (2019)
- [Orbit: A Unified Simulation Framework for Interactive Robot Learning Environments](https://mlsanity.com/paper/2301.04195/) (2023)
- [Sequential Neural Likelihood: Fast Likelihood-free Inference with Autoregressive Flows](https://mlsanity.com/paper/1805.07226/) (2018)
- [Automatic Posterior Transformation for Likelihood-Free Inference](https://mlsanity.com/paper/1905.07488/) (2019)
- [Flexible statistical inference for mechanistic models of neural dynamics](https://mlsanity.com/paper/1711.01861/) (2017)
- [Modern Bayesian Experimental Design](https://mlsanity.com/paper/2302.14545/) (2023)
- [BayesSim: adaptive domain randomization via probabilistic inference for robotics simulators](https://mlsanity.com/paper/1906.01728/) (2019)
- [Likelihood-free inference by ratio estimation](https://mlsanity.com/paper/1611.10242/) (2016)
- [Real-time gravitational-wave science with neural posterior estimation](https://mlsanity.com/paper/2106.12594/) (2021)
- [Learning Rope Manipulation Policies Using Dense Object Descriptors Trained on Synthetic Depth Data](https://mlsanity.com/paper/2003.01835/) (2020)
- [Robust Neural Posterior Estimation and Statistical Model Criticism](https://mlsanity.com/paper/2210.06564/) (2022)
- [ASID: Active Exploration for System Identification in Robotic Manipulation](https://mlsanity.com/paper/2404.12308/) (2024)
- [Sequential Bayesian Experimental Design for Implicit Models via Mutual Information](https://mlsanity.com/paper/2003.09379/) (2020)
- [Learning to See Physical Properties with Active Sensing Motor Policies](https://mlsanity.com/paper/2311.01405/) (2023)
- [Likelihood-free inference of experimental Neutrino Oscillations using Neural Spline Flows](https://mlsanity.com/paper/2002.09436/) (2020)
- [Push to know! -- Visuo-Tactile based Active Object Parameter Inference with Dual Differentiable Filtering](https://mlsanity.com/paper/2308.01001/) (2023)
- [Towards Effective and Efficient Adversarial Defense with Diffusion Models for Robust Visual Tracking](https://mlsanity.com/paper/2506.00325/) (2025)
- [Towards Adaptive Meta-Gradient Adversarial Examples for Visual Tracking](https://mlsanity.com/paper/2505.08999/) (2025)

## Abstract (arXiv)

Black-box simulators are widely used in robotics, but optimizing their parameters remains challenging due to inaccessible likelihoods. Simulation-Based Inference (SBI) tackles this issue using simulation-driven approaches, estimating the posterior from offline real observations and forward simulations. However, in black-box scenarios, preparing observations that contain sufficient information for parameter estimation is difficult due to the unknown relationship between parameters and observations. In this work, we present Active Simulation-Based Inference (ASBI), a parameter estimation framework that uses robots to actively collect real-world online data to achieve accurate black-box simulator tuning. Our framework optimizes robot actions to collect informative observations by maximizing information gain, which is defined as the expected reduction in Shannon entropy between the posterior and the prior. While calculating information gain requires the likelihood, which is inaccessible in black-box simulators, our method solves this problem by leveraging Neural Posterior Estimation (NPE), which leverages a neural network to learn the posterior estimator. Three simulation experiments quantitatively verify that our method achieves accurate parameter estimation, with posteriors sharply concentrated around the true parameters. Moreover, we show a practical application using a real robot to estimate the simulation parameters of cubic particles corresponding to two real objects, beads and gravel, with a bucket pouring action.