PAC-Bayes generalization certificates for learned inductive conformal prediction
Abstract Inductive Conformal Prediction (ICP) provides a practical and effective approach for
equipping deep learning models with uncertainty estimates in the form of set-valued …
equipping deep learning models with uncertainty estimates in the form of set-valued …
Sim-to-lab-to-real: Safe reinforcement learning with shielding and generalization guarantees
Safety is a critical component of autonomous systems and remains a challenge for learning-
based policies to be utilized in the real world. In particular, policies learned using …
based policies to be utilized in the real world. In particular, policies learned using …
Invariant policy optimization: Towards stronger generalization in reinforcement learning
A Sonar, V Pacelli, A Majumdar - Learning for Dynamics …, 2021 - proceedings.mlr.press
A fundamental challenge in reinforcement learning is to learn policies that generalize
beyond the operating domains experienced during training. In this paper, we approach this …
beyond the operating domains experienced during training. In this paper, we approach this …
Learning latent representations to co-adapt to humans
When robots interact with humans in homes, roads, or factories the human's behavior often
changes in response to the robot. Non-stationary humans are challenging for robot learners …
changes in response to the robot. Non-stationary humans are challenging for robot learners …
Task-driven out-of-distribution detection with statistical guarantees for robot learning
Our goal is to perform out-of-distribution (OOD) detection, ie, to detect when a robot is
operating in environments that are drawn from a different distribution than the environments …
operating in environments that are drawn from a different distribution than the environments …
Failure prediction with statistical guarantees for vision-based robot control
We are motivated by the problem of performing failure prediction for safety-critical robotic
systems with high-dimensional sensor observations (eg, vision). Given access to a black …
systems with high-dimensional sensor observations (eg, vision). Given access to a black …
Learning differentiable safety-critical control using control barrier functions for generalization to novel environments
H Ma, B Zhang, M Tomizuka… - 2022 European Control …, 2022 - ieeexplore.ieee.org
Control barrier functions (CBFs) have become a popular tool to enforce safety of a control
system. CBFs are commonly utilized in a quadratic program formulation (CBF-QP) as safety …
system. CBFs are commonly utilized in a quadratic program formulation (CBF-QP) as safety …
Safe perception-based control under stochastic sensor uncertainty using conformal prediction
We consider perception-based control using state estimates that are obtained from high-
dimensional sensor measurements via learning-enabled perception maps. However, these …
dimensional sensor measurements via learning-enabled perception maps. However, these …
A PAC-Bayesian Framework for Optimal Control with Stability Guarantees
Stochastic Nonlinear Optimal Control (SNOC) involves minimizing a cost function that
averages out the random uncertainties affecting the dynamics of nonlinear systems. For …
averages out the random uncertainties affecting the dynamics of nonlinear systems. For …
Integrative use of computer vision and unmanned aircraft technologies in public inspection: Foreign object debris image collection
Unmanned Aircraft Systems (UAS) have become an important resource for public service
providers and smart cities. The purpose of this study is to expand this research area by …
providers and smart cities. The purpose of this study is to expand this research area by …