English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

Variational Neural Belief Parameterizations for Robust Dexterous Grasping via Differentiable CVaR Optimization

Forum topic · 小凯 · 2026-04-30

Summary

This paper addresses stochastic grasp execution caused by contact variability, sensing uncertainty, and external disturbances. Standard expected-quality objectives ignore tail outcomes, causing risk-sensitive POMDPs to select grasps that fail under adverse contact realizations. The authors reformulate grasp acquisition as variational inference over latent contact parameters and object pose, representing the belief as a differentiable Gaussian mixture. They apply Gumbel-Softmax component selection and location-scale reparameterization so that samples become smooth functions of belief parameters, enabling pathwise gradients through a differentiable Conditional Value-at-Risk (CVaR) surrogate for direct optimization of tail robustness. In simulation, the variational neural belief improves robust grasp success under contact and force disturbances and reduces planning time by roughly an order of magnitude compared to particle-filter model-predictive control. Validated on a serial-chain arm and multi-finger hand, the controller terminates in fewer steps and shorter wall-clock time, achieves higher tactile grasp-quality proxies, and produces better-calibrated risk estimates, keeping mean absolute calibration error below 0.14 versus 0.58 for a cross-entropy-method planner.

Overview

  • Field: Machine Learning / Robotics
  • Authors: Clinton Enwerem, Shreya Kalyanaraman, John S. Baras, et al.
  • Released: 2026-04-29
  • arXiv: 2504.21123
  • Key Points

  • Problem: Grasp execution is stochastic due to contact variability, sensor noise, and external disturbances. Expected-quality objectives ignore tail outcomes and often select grasps that fail under adverse contact realizations.
  • Limitation of prior work: Risk-sensitive POMDPs address this failure mode, but many rely on particle-filter beliefs that scale poorly, block gradient-based optimization, and yield high-variance approximations of Conditional Value-at-Risk (CVaR).
  • Approach: The authors reformulate grasp acquisition as variational inference over latent contact parameters and object pose. The belief is represented by a differentiable Gaussian mixture.
  • Reparameterization tricks: Gumbel-Softmax component selection and location-scale reparameterization express samples as smooth functions of belief parameters.
  • Optimization: A differentiable CVaR surrogate allows pathwise gradients, enabling direct optimization of tail robustness.
  • Simulation results: The variational neural belief improves robust grasp success under contact-parameter uncertainty and external-force disturbances, while reducing planning time by roughly an order of magnitude versus particle-filter model-predictive control.
  • Hardware validation: On a serial-chain arm with a multi-finger hand, the proposed controller matches a Gaussian baseline in success under object-pose uncertainty, but terminates in fewer steps with shorter wall-clock time and yields higher tactile grasp-quality proxies.
  • Calibration: The learned belief is better calibrated; mean absolute calibration error stays below 0.14, compared to 0.58 for a cross-entropy-method planner.

Source Excerpt (English Abstract)

> Contact variability, sensing uncertainty, and external disturbances make grasp execution stochastic. Expected-quality objectives ignore tail outcomes and often select grasps that fail under adverse contact realizations. Risk-sensitive POMDPs address this failure mode, but many use particle-filter beliefs that scale poorly, obstruct gradient-based optimization, and estimate Conditional Value-at-Risk (CVaR) with high-variance approximations. We instead formulate grasp acquisition as variational inference over latent contact parameters and object pose, representing the belief with a differentiable Gaussian mixture. We use Gumbel-Softmax component selection and location-scale reparameterization to express samples as smooth functions of the belief parameters, enabling pathwise gradients through...

Tags

#robotics#grasp-planning#variational-inference#pomdp#cvar#risk-sensitive-optimization#dexterous-manipulation#arxiv-2504-21123

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/177618922