662 citations
- ETH ZurichCH37 papers
- IBM (United States)US14 papers
- École Polytechnique Fédérale de LausanneCH9 papers
- University of BaselCH7 papers
- IBM Research - Thomas J. Watson Research CenterUS6 papers
- Centre National de la Recherche ScientifiqueFR5 papers
- Universidade de Santiago de CompostelaES5 papers
- University of ZurichCH5 papers
- Center for Research in Molecular Medicine and Chronic DiseasesES4 papers
- Chalmers University of TechnologySE4 papers
- Forschungszentrum JülichDE4 papers
- Paul Scherrer InstituteCH4 papers
7 papers · 1 filter
A Composable Channel-Adaptive Architecture for Seizure Classification
Francesco Carzaniga, Michael Hersche, Kaspar Schindler +1
Objective: We develop a channel-adaptive (CA) architecture that seamlessly processes multi-variate time-series with an arbitrary number of channels, and in particular intracranial…
If Concept Bottlenecks are the Question, are Foundation Models the Answer?
Nicola Debole, Pietro Barbiero, Francesco Giannini +3
Concept Bottleneck Models (CBMs) are neural networks designed to conjoin high performance with ante-hoc interpretability. CBMs work by first mapping inputs (e.g., images) to high-l…
Local Off-Grid Weather Forecasting with Multi-Modal Earth Observation Data
Qidong Yang, Jonathan Giezendanner, Daniel Salles Civitarese +8
Urgent applications like wildfire management and renewable energy generation require precise, localized weather forecasts near the Earth's surface. However, forecasts produced by m…
Fusing Modalities by Multiplexed Graph Neural Networks for Outcome Prediction in Tuberculosis
Niharika S. D'Souza, Hongzhi Wang, Andrea Giovannini +4
In a complex disease such as tuberculosis, the evidence for the disease and its evolution may be present in multiple modalities such as clinical, genomic, or imaging data. Effectiv…
Generalized Key-Value Memory to Flexibly Adjust Redundancy in Memory-Augmented Networks
Denis Kleyko, Geethan Karunaratne, Jan M. Rabaey +2
Memory-augmented neural networks enhance a neural network with an external key-value memory whose complexity is typically dominated by the number of support vectors in the key memo…
Privacy is What We Care About: Experimental Investigation of Federated Learning on Edge Devices
Anirban Das, Thomas Brunschwiler
Federated Learning enables training of a general model through edge devices without sending raw data to the cloud. Hence, this approach is attractive for digital health application…