References of "Lee, Chul Min 50040505"
     in
Bookmark and Share    
Full Text
Peer Reviewed
See detailFederated Learning for Credit Risk Assessment
Lee, Chul Min UL; Delgado Fernandez, Joaquin UL; Potenciano Menci, Sergio UL et al

in Proceedings of the 56th Hawaii International Conference on System Sciences (2023, January 03)

Credit risk assessment is a standard procedure for financial institutions (FIs) when estimating their credit risk exposure. It involves the gathering and processing quantitative and qualitative datasets ... [more ▼]

Credit risk assessment is a standard procedure for financial institutions (FIs) when estimating their credit risk exposure. It involves the gathering and processing quantitative and qualitative datasets to estimate whether an individual or entity will be able to make future required payments. To ensure effective processing of this data, FIs increasingly use machine learning methods. Large FIs often have more powerful models as they can access larger datasets. In this paper, we present a Federated Learning prototype that allows smaller FIs to compete by training in a cooperative fashion a machine learning model which combines key data derived from several smaller datasets. We test our prototype on an historical mortgage dataset and empirically demonstrate the benefits of Federated Learning for smaller FIs. We conclude that smaller FIs can expect a significant performance increase in their credit risk assessment models by using collaborative machine learning. [less ▲]

Detailed reference viewed: 101 (31 UL)
Full Text
Peer Reviewed
See detailPrivacy-preserving federated learning for residential short-term load forecasting
Delgado Fernandez, Joaquin UL; Potenciano Menci, Sergio UL; Lee, Chul Min UL et al

in Applied Energy (2022), 326

With high levels of intermittent power generation and dynamic demand patterns, accurate forecasts for residential loads have become essential. Smart meters can play an important role when making these ... [more ▼]

With high levels of intermittent power generation and dynamic demand patterns, accurate forecasts for residential loads have become essential. Smart meters can play an important role when making these forecasts as they provide detailed load data. However, using smart meter data for load forecasting is challenging due to data privacy requirements. This paper investigates how these requirements can be addressed through a combination of federated learning and privacy preserving techniques such as differential privacy and secure aggregation. For our analysis, we employ a large set of residential load data and simulate how different federated learning models and privacy preserving techniques affect performance and privacy. Our simulations reveal that combining federated learning and privacy preserving techniques can secure both high forecasting accuracy and near-complete privacy. Specifically, we find that such combinations enable a high level of information sharing while ensuring privacy of both the processed load data and forecasting models. Moreover, we identify and discuss challenges of applying federated learning, differential privacy and secure aggregation for residential short-term load forecasting. [less ▲]

Detailed reference viewed: 82 (23 UL)