Trustworthiness Evaluation and Trust-Aware Design of CNN Architectures

Mingxi Cheng, Tingyang Sun, Shahin Nazarian, Paul Bogdan
Proceedings of The 1st Conference on Lifelong Learning Agents, PMLR 199:1086-1102, 2022.

Abstract

Convolutional neural networks (CNNs) are known to be effective tools in many deep learning application areas. Despite CNN’s good performance in terms of classical evaluation metrics such as accuracy and loss, quantifying and ensuring a high degree of trustworthiness of such models remains an unsolved problem raising questions in applications where trust is an important factor. In this work, we propose a framework to evaluate the trustworthiness of CNNs. Towards this end, we develop a trust-based pooling layer for CNNs to achieve higher accuracy and trustworthiness in applications with noise in input features. We further propose TrustCNets consisting of trustworthiness-aware CNN building blocks, i.e., one or more conv layers followed by a trust-based pooling layer. TrustCNets can stack together as a trust-aware CNN architecture or be plugged into deep learning architectures to improve performance. In our experiments, we evaluate the trustworthiness of popular CNN building blocks and demonstrate the performance of our TrustCNet empirically with multiple datasets.

Cite this Paper

BibTeX
@InProceedings{pmlr-v199-cheng22a, title = {Trustworthiness Evaluation and Trust-Aware Design of CNN Architectures}, author = {Cheng, Mingxi and Sun, Tingyang and Nazarian, Shahin and Bogdan, Paul}, booktitle = {Proceedings of The 1st Conference on Lifelong Learning Agents}, pages = {1086--1102}, year = {2022}, editor = {Chandar, Sarath and Pascanu, Razvan and Precup, Doina}, volume = {199}, series = {Proceedings of Machine Learning Research}, month = {22--24 Aug}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v199/cheng22a/cheng22a.pdf}, url = {https://proceedings.mlr.press/v199/cheng22a.html}, abstract = {Convolutional neural networks (CNNs) are known to be effective tools in many deep learning application areas. Despite CNN’s good performance in terms of classical evaluation metrics such as accuracy and loss, quantifying and ensuring a high degree of trustworthiness of such models remains an unsolved problem raising questions in applications where trust is an important factor. In this work, we propose a framework to evaluate the trustworthiness of CNNs. Towards this end, we develop a trust-based pooling layer for CNNs to achieve higher accuracy and trustworthiness in applications with noise in input features. We further propose TrustCNets consisting of trustworthiness-aware CNN building blocks, i.e., one or more conv layers followed by a trust-based pooling layer. TrustCNets can stack together as a trust-aware CNN architecture or be plugged into deep learning architectures to improve performance. In our experiments, we evaluate the trustworthiness of popular CNN building blocks and demonstrate the performance of our TrustCNet empirically with multiple datasets.} }
Endnote
%0 Conference Paper %T Trustworthiness Evaluation and Trust-Aware Design of CNN Architectures %A Mingxi Cheng %A Tingyang Sun %A Shahin Nazarian %A Paul Bogdan %B Proceedings of The 1st Conference on Lifelong Learning Agents %C Proceedings of Machine Learning Research %D 2022 %E Sarath Chandar %E Razvan Pascanu %E Doina Precup %F pmlr-v199-cheng22a %I PMLR %P 1086--1102 %U https://proceedings.mlr.press/v199/cheng22a.html %V 199 %X Convolutional neural networks (CNNs) are known to be effective tools in many deep learning application areas. Despite CNN’s good performance in terms of classical evaluation metrics such as accuracy and loss, quantifying and ensuring a high degree of trustworthiness of such models remains an unsolved problem raising questions in applications where trust is an important factor. In this work, we propose a framework to evaluate the trustworthiness of CNNs. Towards this end, we develop a trust-based pooling layer for CNNs to achieve higher accuracy and trustworthiness in applications with noise in input features. We further propose TrustCNets consisting of trustworthiness-aware CNN building blocks, i.e., one or more conv layers followed by a trust-based pooling layer. TrustCNets can stack together as a trust-aware CNN architecture or be plugged into deep learning architectures to improve performance. In our experiments, we evaluate the trustworthiness of popular CNN building blocks and demonstrate the performance of our TrustCNet empirically with multiple datasets.
APA
Cheng, M., Sun, T., Nazarian, S. & Bogdan, P.. (2022). Trustworthiness Evaluation and Trust-Aware Design of CNN Architectures. Proceedings of The 1st Conference on Lifelong Learning Agents, in Proceedings of Machine Learning Research 199:1086-1102 Available from https://proceedings.mlr.press/v199/cheng22a.html.

Related Material