Analysis and Mitigation of Unwanted Biases in ML-based QoT Classification Tasks
Paper in proceeding, 2024
We address the problem of mitigating biases in models used for the quality of transmission prediction. The proposed method reduces the relative accuracy difference between samples with different feature values by up to 45%.
Optical networks
Machine learning
Classification
Quality of transmission
Author
Carlos Natalino Da Silva
Chalmers, Electrical Engineering, Communication, Antennas and Optical Networks
Behnam Shariati
Fraunhofer Society
Pooyan Safari
Fraunhofer Society
Johannes Fischer
Fraunhofer Society
Paolo Monti
Chalmers, Electrical Engineering, Communication, Antennas and Optical Networks
Conference on Optical Fiber Communication, Technical Digest Series
M1H.3
9781957171326 (ISBN)
Optical Fiber Communications Conference and Exhibition (OFC)
San Diego, CA, USA,
San Diego, CA, USA,
Providing Resilient & secure networks [Operating on Trusted Equipment] to CriTical infrastructures (PROTECT)
VINNOVA (2020-03506), 2021-02-01 -- 2024-01-31.
Areas of Advance
Information and Communication Technology
Subject Categories (SSIF 2011)
Telecommunications
Communication Systems
Signal Processing
DOI
10.1364/OFC.2024.M1H.3