Unfair Photonics? How Encoding May Lead to Bias in Photonic Neural Networks
In this paper, we show that complex-valued Photonic Neural Networks (PNNs) may yield unfair decisions based on encoding strategy and feature pairing, even when these networks are trained to similar accuracy levels.
Authors
- Fabio Pavanello (ORCID: https://orcid.org/0000-0002-6889-2012)
- Paul Jiménez (ORCID: https://orcid.org/0000-0002-8229-1141)
- Maurício Gomes de Queiroz
- Ian O’Connor (ORCID: https://orcid.org/0000-0002-6238-9600)
- Alberto Bosio (ORCID: https://orcid.org/0000-0001-6116-7339)
- Mateus Vidaletti Costa (ORCID: https://orcid.org/0000-0002-7248-8493)
Institutions
- Institut polytechnique de Grenoble (FR)
- Institut des Nanotechnologies de Lyon (FR)
- Université Savoie Mont Blanc (FR)
- RMIT University (AU)
Publication Details
- Published
- 2025-01-01
- DOI
- https://doi.org/10.1364/cleo_at.2025.jps100_106
- Primary Topic
- Neural Networks and Reservoir Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.00