Unlocking the black box: an in-depth review on interpretability, explainability, and reliability in deep learning
Deep learning models have revolutionized numerous fields, yet their decision-making
processes often remain opaque, earning them the characterization of “black-box” models …
processes often remain opaque, earning them the characterization of “black-box” models …
Applications of deep learning for drug discovery systems with bigdata
Y Matsuzaka, R Yashiro - BioMedInformatics, 2022 - mdpi.com
The adoption of “artificial intelligence (AI) in drug discovery”, where AI is used in the process
of pharmaceutical research and development, is progressing. By using the ability to process …
of pharmaceutical research and development, is progressing. By using the ability to process …
Res: A robust framework for guiding visual explanation
Despite the fast progress of explanation techniques in modern Deep Neural Networks
(DNNs) where the main focus is handling" how to generate the explanations", advanced …
(DNNs) where the main focus is handling" how to generate the explanations", advanced …
Aligning eyes between humans and deep neural network through interactive attention alignment
While Deep Neural Networks (DNNs) are deriving the major innovations through their
powerful automation, we are also witnessing the peril behind automation as a form of bias …
powerful automation, we are also witnessing the peril behind automation as a form of bias …
On" Deep Learning" Misconduct
J Weng - arXiv preprint arXiv:2211.16350, 2022 - arxiv.org
This is a theoretical paper, as a companion paper of the plenary talk for the same
conference ISAIC 2022. In contrast to the author's plenary talk in the same conference …
conference ISAIC 2022. In contrast to the author's plenary talk in the same conference …
Saliency-regularized deep multi-task learning
Multi-task learning (MTL) is a framework that enforces multiple learning tasks to share their
knowledge to improve their generalization abilities. While shallow multi-task learning can …
knowledge to improve their generalization abilities. While shallow multi-task learning can …
Misconduct in Post-Selections and Deep Learning
J Weng - 2023 8th International Conference on Control …, 2024 - ieeexplore.ieee.org
This is a theoretical paper on" Deep Learning" misconduct in particular and Post-Selection
in general. As far as the author knows, the first peer-reviewed papers on Deep Learning …
in general. As far as the author knows, the first peer-reviewed papers on Deep Learning …
The luckiest network gives the average error on disjoint tests: Experiments
X Wu, J Weng - Proceedings of the 2024 5th International Conference …, 2024 - dl.acm.org
This is an experimental paper associated with the theoretical paper Weng [34] addressing
the issue of “Deep Learning” misconduct in particular and Post-Selection in general …
the issue of “Deep Learning” misconduct in particular and Post-Selection in general …
Schematic memory persistence and transience for efficient and robust continual learning
Continual learning is considered a promising step toward next-generation Artificial
Intelligence (AI), where deep neural networks (DNNs) make decisions by continuously …
Intelligence (AI), where deep neural networks (DNNs) make decisions by continuously …
Benchmarking the Effect of Poisoning Defenses on the Security and Bias of Deep Learning Models
Machine learning models are susceptible to a class of attacks known as adversarial
poisoning where an adversary can maliciously manipulate training data to hinder model …
poisoning where an adversary can maliciously manipulate training data to hinder model …