TY - GEN
T1 - Out of Distribution Detection, Generalization, and Robustness Triangle with Maximum Probability Theorem
AU - Marvasti, Amir Emad
AU - Marvasti, Ehsan Emad
AU - Bagci, Ulas
N1 - Funding Information:
This project is supported by the NIH funding: R01-CA246704 and R01-CA240639, and Florida Department of Health (FDOH): 20K04.
Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - Maximum Probability Framework, powered by Maximum Probability Theorem, is a recent theoretical development in artificial intelligence, aiming to formally define probabilistic models, guiding development of objective functions, and regularization of probabilistic models. MPT uses the probability distribution that the models assume on random variables to provide an upper bound on the probability of the model. We apply MPT to challenging out-of-distribution (OOD) detection problems in computer vision by incorporating MPT as a regularization scheme in the training of CNNs and their energy-based variants. We demonstrate the effectiveness of the proposed method on 1080 trained models, with varying hyperparameters, and conclude that the MPT-based regularization strategy stabilizes and improves the generalization and robustness of base models in addition to enhanced OOD performance on CIFAR10, CIFAR100, and MNIST datasets.
AB - Maximum Probability Framework, powered by Maximum Probability Theorem, is a recent theoretical development in artificial intelligence, aiming to formally define probabilistic models, guiding development of objective functions, and regularization of probabilistic models. MPT uses the probability distribution that the models assume on random variables to provide an upper bound on the probability of the model. We apply MPT to challenging out-of-distribution (OOD) detection problems in computer vision by incorporating MPT as a regularization scheme in the training of CNNs and their energy-based variants. We demonstrate the effectiveness of the proposed method on 1080 trained models, with varying hyperparameters, and conclude that the MPT-based regularization strategy stabilizes and improves the generalization and robustness of base models in addition to enhanced OOD performance on CIFAR10, CIFAR100, and MNIST datasets.
KW - deep learning
KW - maximum probability theorem
KW - Out of distribution detection
KW - regularization
KW - robustness
UR - http://www.scopus.com/inward/record.url?scp=85146421176&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85146421176&partnerID=8YFLogxK
U2 - 10.1109/ICECCME55909.2022.9988128
DO - 10.1109/ICECCME55909.2022.9988128
M3 - Conference contribution
AN - SCOPUS:85146421176
T3 - International Conference on Electrical, Computer, Communications and Mechatronics Engineering, ICECCME 2022
BT - International Conference on Electrical, Computer, Communications and Mechatronics Engineering, ICECCME 2022
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2022 International Conference on Electrical, Computer, Communications and Mechatronics Engineering, ICECCME 2022
Y2 - 16 November 2022 through 18 November 2022
ER -
