TY  - GEN
PB  - Institute of Electrical and Electronics Engineers (IEEE)
UR  - https://doi.org/10.1109/WACV48630.2021
ID  - discovery10132169
N2  - In this paper, we propose risk-calibrated evidential deep classifiers to reduce the costs associated with classification errors. We use two main approaches. The first is to develop methods to quantify the uncertainty of a classifier?s predictions and reduce the likelihood of acting on erroneous predictions. The second is a novel way to train the classifier such that erroneous classifications are biased towards less risky categories. We combine these two approaches in a principled way. While doing this, we extend evidential deep learning with pignistic probabilities, which are used to quantify uncertainty of classification predictions and model rational decision making under uncertainty.We evaluate the performance of our approach on several image classification tasks. We demonstrate that our approach allows to (i) incorporate misclassification cost while training deep classifiers, (ii) accurately quantify the uncertainty of classification predictions, and (iii) simultaneously learn how to make classification decisions to minimize expected cost of classification errors.
KW  - Training
KW  -  Deep learning
KW  -  Computer vision
KW  - 
Uncertainty
KW  -  Conferences
KW  -  Decision making
KW  -  Predictive models
A1  - Sensoy, M
A1  - Saleki, M
A1  - Julier, S
A1  - Aydogan, R
A1  - Reid, J
EP  - 2491
Y1  - 2021///
AV  - public
SP  - 2483
TI  - Misclassification Risk and Uncertainty Quantification in Deep Classifiers
N1  - This version is the author accepted manuscript. For information on re-use, please refer to the publisher's terms and conditions.
ER  -