Published in 36th Neurips (Spotlight), 2024
Our paper proposes a mechanism design model augmented with an output prediction and applies it to various problems using a universal error measure in the algorithmic design with predictions setting.
Recommended citation: Christodoulou George, Sgouritsa Alkmini and Vlachos Ioannis (2024). "Mechanism Design Augmented with Output Advice." arXiv preprint arXiv:2406.14165.
Download Paper | Download Slides
Published in Trustworthy AI - Integrating Learning, Optimization and Reasoning, 2021
A novel methodology is presented, in which deep neural architectures that have been trained to provide highly accurate predictions over existing datasets are adapted, in a consistent way, to make predictions over different contexts and datasets.
Recommended citation: D. Kollias, Y. Vlaxos, M. Seferis, I. Kollia, L. Sukissian, J. Wingate, and S. Kollias. (2021). "Transparent adaptation in deep medical image diagnosis." Trustworthy AI - Integrating Learning, Optimization and Reasoning.
Download Paper
Published in arXiv preprint, 2020
The paper presents a novel deep learning approach, which extracts latent information from trained Deep Neural Networks (DNNs) and derives concise representations that are analyzed in an effective, unified way for prediction purposes.
Recommended citation: D. Kollias, N. Bouas, Y. Vlaxos, V. Brillakis, M. Seferis, I. Kollia, L. Sukissian, J. Wingate, and S. Kollias. (2020). "Deep Transparent Prediction through Latent Representation Analysis." arXiv:2009.07044.
Download Paper