ABSTRACT
Patient-centric adaptation of audiological preferences across different contexts is a challenging task, as traditional clinical measurements of audibility do not reflect the cognitive perception of speech nor binaural loudness of sounds in different contexts. Smartphone-based machine learning personalization systems have the potential to address this issue in real-world listening scenarios, however, the necessary training datasets are not currently available. As hearing healthcare medical data is of a highly private nature, a framework is proposed, combining federated learning (FL) and secret sharing in the context of hearing aids with the goal of training models locally while preserving the individual user's privacy. We demonstrate an application of such a system with a simplified domain defined by the MNIST digit classification task.
- Lopez-Poveda, E. 2014. Why do I hear but not understand? Stochastic undersampling as a model of degraded neural encoding of speech. Frontiers in Neuroscience. 8, (2014).Google Scholar
- Oetting, D. et al. 2018. Restoring Perceived Loudness for Listeners With Hearing Loss. Ear and Hearing. 39, 4 (2018), 664--678.Google ScholarCross Ref
- Korzepa, M. et al. 2018. Learning Preferences and Soundscapes for Augmented Hearing. CEUR Workshop Proceedings. 2068, (2018).Google Scholar
- Pasta, A. et al. 2019. Rethinking Hearing Aids as Recommender Systems. ACM Proceedings of RecSys. (2019).Google Scholar
- Rieke, N. et al. 2020. The Future of Digital Health with Federated Learning. arXiv. 2003.08119, (2020).Google Scholar
- Konecný, J. et al. 2016. Federated Optimization: Distributed Machine Learning for On-Device Intelligence. arXiv. 1610.02527, (2016).Google Scholar
- Lesica, N. 2018. Why Do Hearing Aids Fail to Restore Normal Auditory Perception?. Trends in Neurosciences. 41, 4 (2018), 174--185.Google ScholarCross Ref
- Sanchez Lopez, R. et al. 2018. Data-Driven Approach for Auditory Profiling and Characterization of Individual Hearing Loss. Trends in Hearing. 22, (2018), 233121651880740.Google Scholar
- Sweeney, L. et al. 2013. Identifying Participants in the Personal Genome Project by Name. SSRN Electronic Journal. (2013).Google Scholar
- Besser, J. et al. 2018. Comorbidities of hearing loss and the implications of multimorbidity for audiological care. Hearing Research. 369, (2018), 3--14.Google Scholar
- He, K. et al. 2015. Deep Residual Learning for Image Recognition. arXiv. 1512.03385, (2015).Google Scholar
- Bhagoji, A. et al. 2019. Analyzing Federated Learning through an Adversarial Lens. arXiv. 1811.12470, (2019).Google Scholar
- Yang, T. et al. 2020. Efficient Deep Reinforcement Learning through Policy Transfer. arXiv. 2002.08037, (2020).Google Scholar
- Yin, H. and Pan, S. 2017. Knowledge Transfer for Deep Reinforcement Learning with Hierarchical Experience Replay. AAAI-17 (San Francisco, 2017), 1640--1646.Google Scholar
Index Terms
- Modelling Audiological Preferences using Federated Learning
Recommendations
Measuring and modeling context-dependent preferences for hearing aid settings
AbstractDespite having individual perceptual preferences toward sounds, hearing aid users often end up with default hearing aid settings that have no contextual awareness. However, the introduction of smartphone-connected hearing aids has enabled a ...
PerFedRec++: Enhancing Personalized Federated Recommendation with Self-Supervised Pre-Training
Federated recommendation systems employ federated learning techniques to safeguard user privacy by transmitting model parameters instead of raw user data between user devices and the central server. Nevertheless, the current federated recommender system ...
New Recommendation Techniques for Multicriteria Rating Systems
Traditional single-rating recommender systems have been successful in a number of personalization applications, but the research area of multicriteria recommender systems has been largely untouched. Taking full advantage of multicriteria ratings in ...
Comments