Data-efficient, Explainable and Safe Box Manipulation: Illustrating the Advantages of Physical Priors in Model-Predictive Control

• URL: http://arxiv.org/abs/2303.01563v2
• Date: Thu, 28 Mar 2024 21:16:15 GMT
• Title: Data-efficient, Explainable and Safe Box Manipulation: Illustrating the Advantages of Physical Priors in Model-Predictive Control
• Authors: Achkan Salehi, Stephane Doncieux,
• Abstract summary: We show that prior knowledge of environment dynamics in an MPC framework can lead to improvements in explainability, safety and data-efficiency. We model a payload manipulation problem based on a real robotic system, and show that leveraging prior knowledge about the dynamics of the environment in an MPC framework can lead to improvements in explainability, safety and data-efficiency.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Model-based RL/control have gained significant traction in robotics. Yet, these approaches often remain data-inefficient and lack the explainability of hand-engineered solutions. This makes them difficult to debug/integrate in safety-critical settings. However, in many systems, prior knowledge of environment kinematics/dynamics is available. Incorporating such priors can help address the aforementioned problems by reducing problem complexity and the need for exploration, while also facilitating the expression of the decisions taken by the agent in terms of physically meaningful entities. Our aim with this paper is to illustrate and support this point of view via a case-study. We model a payload manipulation problem based on a real robotic system, and show that leveraging prior knowledge about the dynamics of the environment in an MPC framework can lead to improvements in explainability, safety and data-efficiency, leading to satisfying generalization properties with less data.

Related papers

• Harnessing physics-informed operators for high-dimensional reliability analysis problems [0.8192907805418583]
  Reliability analysis is a formidable task, particularly in systems with a large number of parameters. Conventional methods for quantifying reliability often rely on extensive simulations or experimental data. We show that physics-informed operator can seamlessly solve high-dimensional reliability analysis problems with reasonable accuracy.
  arXiv  Detail & Related papers  (2024-09-07T04:52:03Z)
• Analyzing Adversarial Inputs in Deep Reinforcement Learning [53.3760591018817]
  We present a comprehensive analysis of the characterization of adversarial inputs, through the lens of formal verification. We introduce a novel metric, the Adversarial Rate, to classify models based on their susceptibility to such perturbations. Our analysis empirically demonstrates how adversarial inputs can affect the safety of a given DRL system with respect to such perturbations.
  arXiv  Detail & Related papers  (2024-02-07T21:58:40Z)
• Effective Communication with Dynamic Feature Compression [25.150266946722]
  We study a prototypal system in which an observer must communicate its sensory data to a robot controlling a task. We consider an ensemble Vector Quantized Variational Autoencoder (VQ-VAE) encoding, and train a Deep Reinforcement Learning (DRL) agent to dynamically adapt the quantization level. We tested the proposed approach on the well-known CartPole reference control problem, obtaining a significant performance increase.
  arXiv  Detail & Related papers  (2024-01-29T15:35:05Z)
• A spectrum of physics-informed Gaussian processes for regression in engineering [0.0]
  Despite the growing availability of sensing and data in general, we remain unable to fully characterise many in-service engineering systems and structures from a purely data-driven approach. This paper pursues the combination of machine learning technology and physics-based reasoning to enhance our ability to make predictive models with limited data.
  arXiv  Detail & Related papers  (2023-09-19T14:39:03Z)
• On Robust Numerical Solver for ODE via Self-Attention Mechanism [82.95493796476767]
  We explore training efficient and robust AI-enhanced numerical solvers with a small data size by mitigating intrinsic noise disturbances. We first analyze the ability of the self-attention mechanism to regulate noise in supervised learning and then propose a simple-yet-effective numerical solver, Attr, which introduces an additive self-attention mechanism to the numerical solution of differential equations.
  arXiv  Detail & Related papers  (2023-02-05T01:39:21Z)
• Denoised MDPs: Learning World Models Better Than the World Itself [94.74665254213588]
  This work categorizes information out in the wild into four types based on controllability and relation with reward, and formulates useful information as that which is both controllable and reward-relevant. Experiments on variants of DeepMind Control Suite and RoboDesk demonstrate superior performance of our denoised world model over using raw observations alone.
  arXiv  Detail & Related papers  (2022-06-30T17:59:49Z)
• How Can Subgroup Discovery Help AIOps? [0.0]
  We study how Subgroup Discovery can help AIOps. This project involves both data mining researchers and practitioners from Infologic, a French software editor.
  arXiv  Detail & Related papers  (2021-09-10T14:41:02Z)
• Counterfactual Explanations as Interventions in Latent Space [62.997667081978825]
  Counterfactual explanations aim to provide to end users a set of features that need to be changed in order to achieve a desired outcome. Current approaches rarely take into account the feasibility of actions needed to achieve the proposed explanations. We present Counterfactual Explanations as Interventions in Latent Space (CEILS), a methodology to generate counterfactual explanations.
  arXiv  Detail & Related papers  (2021-06-14T20:48:48Z)
• Offline Model-Based Optimization via Normalized Maximum Likelihood Estimation [101.22379613810881]
  We consider data-driven optimization problems where one must maximize a function given only queries at a fixed set of points. This problem setting emerges in many domains where function evaluation is a complex and expensive process. We propose a tractable approximation that allows us to scale our method to high-capacity neural network models.
  arXiv  Detail & Related papers  (2021-02-16T06:04:27Z)
• Sufficiently Accurate Model Learning for Planning [119.80502738709937]
  This paper introduces the constrained Sufficiently Accurate model learning approach. It provides examples of such problems, and presents a theorem on how close some approximate solutions can be. The approximate solution quality will depend on the function parameterization, loss and constraint function smoothness, and the number of samples in model learning.
  arXiv  Detail & Related papers  (2021-02-11T16:27:31Z)
• Variational Autoencoding of PDE Inverse Problems [12.716429755564821]
  Modern machine learning allows one to circumvent problems involving prior knowledge and physical laws. In this work we fold the mechanistic model into a flexible data-driven surrogate to arrive at a physically structured decoder network. We employ the variational form of the PDE problem and introduce local approximations as a form of model based data augmentation.
  arXiv  Detail & Related papers  (2020-06-28T16:17:03Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.