BriefGPT.xyz
May, 2019
使用显著性图预测自动驾驶系统中的模型故障
Predicting Model Failure using Saliency Maps in Autonomous Driving Systems
HTML
PDF
Sina Mohseni, Akshay Jagadeesh, Zhangyang Wang
TL;DR
该研究提出了一种利用显著性地图训练预测模型错误的学生模型来评估模型健壮性和估计机器学习系统故障概率的设计，并在自动驾驶车辆控制系统等安全关键应用中实现并审查了该故障预测模型的初步结果。
Abstract
While
machine learning
systems show high success rate in many complex tasks, research shows they can also fail in very unexpected situations. Rise of
machine learning
products in safety-critical industries cause
→