Abstract
The application of machine learning to insurance risk prediction requires learning from sensitive data. This raises multiple ethical and legal issues. One of the most relevant ones is privacy. However, privacy-preserving methods can potentially hinder the predictive potential of machine learning models. In this paper, we present preliminary experiments with life insurance data using two privacy-preserving techniques: discretization and encryption. Our objective with this work is to assess the impact of such privacy preservation techniques in the accuracy of ML models. We instantiate the problem in three general, but plausible Use Cases involving the prediction of insurance claims within a 1-year horizon. Our preliminary experiments suggest that discretization and encryption have negligible impact in the accuracy of ML models. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Abstract

Abstract

Abstract
Construction and analysis of distributed systems is difficult. Multiparty session types (MPST) constitute a method to make it easier. The idea is to use type checking to statically prove deadlock freedom and protocol compliance of communicating processes. In practice, the premier approach to apply the MPST method in combination with mainstream programming languages has been based on API generation. In this paper (pearl), we revisit and revise this approach. Regarding our “revisitation”, using Scala 3, we present the existing API generation approach, which is based on deterministic finite automata (DFA), in terms of both the existing states-as-classes encoding of DFAs as APIs, and a new states-as-type-parameters encoding; the latter leverages match types in Scala 3. Regarding our “revision”, also using Scala 3, we present a new API generation approach that is based on sets of pomsets instead of DFAs; it crucially leverages match types, too. Our fresh perspective allows us to avoid two forms of combinatorial explosion resulting from implementing concurrent subprotocols in the DFA-based approach. We implement our approach in a new API generation tool. © Guillermina Cledou, Luc Edixhoven, Sung-Shik Jongmans, and Jos Proena; licensed under Creative Commons License CC-BY 4.0

Abstract

Abstract
Multiparty session types (MPST) constitute a method to simplify construction and analysis of distributed systems. The idea is that well-typedness of processes at compile-time (statically) entails deadlock freedom and protocol compliance of their sessions of communications at execution-time (dynamically). In practice, the premier approach to apply the MPST method in combination with mainstream programming languages has been based on API generation. However, existing MPST tools support only unilingual programming (homogeneity), while many real-world distributed systems are engineered using multilingual programming (heterogeneity). In this paper, we present a blueprint of ST4MP: a tool to apply the MPST method in multilingual programming, based on API generation. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.