Sample-efficient safety assurances using conformal prediction

Rachel Luo

doi:10.1177/02783649231221580

Sample-efficient safety assurances using conformal prediction

2023-12-17

1

OA

AI

DOI：10.1177/02783649231221580

Original

Original request for help

Save

Abstract

En 中文

When deploying machine learning models in high-stakes robotics applications, the ability to detect unsafe situations is crucial. Early warning systems can provide alerts when an unsafe situation is imminent (in the absence of corrective action). To reliably improve safety, these warning systems should have a provable false negative rate; that is, of the situations that are unsafe, fewer than epsilon will occur without an alert. In this work, we present a framework that combines a statistical inference technique known as conformal prediction with a simulator of robot/environment dynamics, in order to tune warning systems to provably achieve an epsilon false negative rate using as few as 1/epsilon data points. We apply our framework to a driver warning system and a robotic grasping application, and empirically demonstrate the guaranteed false negative rate while also observing a low false detection (positive) rate.

Keywords:

Safety assurance

conformal prediction

statistical inference

Sample-efficient safety assurances using conformal prediction

Abstract

Journal

Organization