Abstract
Ordered Weighted Averaging (OWA) operators have been integrated in Convolutional Neural Networks (CNNs) for image classification through the OWA layer. This layer lets the CNN integrate global information about the image in the early stages, where most CNN architectures only allow for the exploitation of local information. As a side effect of this integration, the OWA layer becomes a practical method for the determination of OWA operator weights, which is usually a difficult task that complicates the integration of these operators in other fields. In this paper, we explore the weights learned for the OWA operators inside the OWA layer, characterizing them through their basic properties of orness and dispersion. We also compare them to some families of OWA operators, namely the Binomial OWA operator, the Stancu OWA operator and the exponential RIM OWA operator, finding examples that are currently impossible to generalize through these parameterizations.
Highlights
Ordered Weighted Averaging (OWA) Operators Learned in Information fusion refers to the process of combining different sources of information into a single output, which tries to summarize those inputs into a single element
Aiming at integrating OWA operators in Convolutional Neural Networks (CNNs) to improve the performance of CNNs in image classification problems, in a previous work [13], we introduced the OWA layer
The OWA layer that we explore in this work was previously defined in [13,38], where we presented some preliminary results about the impact of the layer in CNNs
Summary
OWA Operators Learned in Information fusion refers to the process of combining different sources of information into a single output, which tries to summarize those inputs into a single element. OWA operators can generalize a large array of aggregations, such as the minimum, the maximum, or the median, among others This ability to represent very different aggregations has one downside, namely the difficulty of determining the specific weights for an OWA operator in a given context, leading to the development of several methods to establish their parameters [6,7,8,9]. Most of these methods [6] are based on translating the opinion of an expert to a specific weighting vector, usually restricting the natural complexity of the operator to a family of OWAs that are easier to parameterize. The weights are obtained from data [8,9], but even in most of these proposals, the OWA operators are still constrained to parameterized variants
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
