@misc{17750,
  author = {Youcef Djenouri and Ahmed Belbachir and Nassim Belmecheri and Asma Belhadi and Tomasz Michalak},
  title = {Shapley Consensus Deep Learning for Ensemble Pruning},
  abstract = {This paper targets a new foundation for designing
general-purpose learning systems, by establishing a con-
sensus method that facilitates self-adaptation and flexibility
to deal with different learning tasks and different data dis-
tribution. We present the Shapely Consensus Deep Learning
(SCDL) as a consensus method for general-purpose intelli-
gence without the help of a domain expert. SCDL is two-
level based learning process. In the first level, several deep
learning models have been trained for each historical ob-
servation. The Shapley Value is determined to compute the
contribution of each subset of models in the training. The
models are pruned according to their contribution in the
learning process. In the second level, the loss information
of each data distribution is saved in the knowledge base.
Both levels are explored to prune the models for each new
observation. We present the evaluation of the generality
of SCDL using different datasets with different shapes, and
complexities. The results reveal the effectiveness of SCDL
for weakly classification. Concretely, SCDL achieved 90\%
of AUC with less than 86\% for the baseline solutions.},
  year = {2025},
  journal = {IEEE/CVF Winter Conference on Applications of Computer Vision},
  publisher = {IEEE},
}