Design and Evaluation of Object Classifiers for Probabilistic Decision-Making in Autonomous Systems
Design and Evaluation of Object Classifiers for Probabilistic Decision-Making in Autonomous Systems
复制标题
DOI：
10.1109/icra46639.2022.9812171
复制
发表时间：
2022-05
期刊：
影响因子：
--
通讯作者：
Hamad Ullah;Weisi Fan;T. Wongpiromsarn
中科院分区：
文献类型：
--
作者：
Hamad Ullah;Weisi Fan;T. Wongpiromsarn
在许多自治系统中，目标分类是实现有效决策的关键因素。一个更复杂的系统也可以利用类的概率分布，而不是仅仅基于最可能的类做出决策。本文引入了新的性能指标：绝对类误差（ACE）、绝对类误差期望（EACE）和绝对类误差方差（VACE）来评估这些概率的准确性。我们使用不同的神经网络架构和数据集来测试这个度量。在此基础上，提出了一种新的基于任务的目标分类神经网络，并将其性能与典型的概率分类模型进行了比较，以显示基于阈值的概率决策的改进。
Object classification is a key element that enables effective decision-making in many autonomous systems. A more sophisticated system may also utilize the probability distribution over the classes instead of basing its decision only on the most likely class. This paper introduces new performance metrics: the absolute class error (ACE), expectation of absolute class error (EACE) and variance of absolute class error (VACE) for evaluating the accuracy of such probabilities. We test this metric using different neural network architectures and datasets. Furthermore, we present a new task-based neural network for object classification and compare its performance with a typical probabilistic classification model to show the improvement with threshold-based probabilistic decision-making.