![]() Lee, Chul Min ![]() ![]() ![]() 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: 128 (40 UL)![]() Barbereau, Tom Josua ![]() ![]() ![]() in Technology in Society (2023), 73 Ethereum's public distributed ledger can issue tokenised voting rights that are tradable on crypto-asset exchanges by potentially anyone. Ethereum thus enables global, unincorporated associations to ... [more ▼] Ethereum's public distributed ledger can issue tokenised voting rights that are tradable on crypto-asset exchanges by potentially anyone. Ethereum thus enables global, unincorporated associations to conduct governance experiments. Such experiments are crucial to Decentralised Finance (DeFi). DeFi is a nascent field of unlicensed, unregulated, and non-custodial financial services that utilise public distributed ledgers and crypto-assets rather than corporate structures and sovereign currencies. The inaugural Bloomberg Galaxy DeFi Index, launched in August 2021, included nine Ethereum-based projects – non-custodial exchanges as well as lending and derivatives platforms. Each project is governed, at least in part, by unregistered holders of tokenised voting rights (also known as governance tokens). Token-holders typically vote for or against coders' improvement proposals that pertain to anything from the allocation of treasury funds to a collateral's risk parameters. DeFi's governance thus depends on the distribution and exercise of tokenised voting rights. Since archetypal DeFi projects are not managed by companies or public institutions, not much is known about DeFi's governance. Regulators and law-makers from the United States recently asked if DeFi's governance entails a new class of “shadowy” elites. In response, we conducted an exploratory, multiple-case study that focused on the tokenised voting rights issued by the nine projects from Bloomberg's inaugural Galaxy DeFi index. Our mixed methods approach drew on Ethereum-based data about the distribution, trading, staking, and delegation of voting rights tokens, as well as project documentation and archival records. We discovered that DeFi projects' voting rights are highly concentrated, and the exercise of these rights is very low. Our theoretical contribution is a philosophical intervention: minority rule, not “democracy”, is the probable outcome of token-tradable voting rights and a lack of applicable anti-concentration laws. We interpret DeFi's minority rule as timocratic. [less ▲] Detailed reference viewed: 62 (13 UL)![]() ; ; et al Report (2022) Detailed reference viewed: 45 (5 UL)![]() ; ; et al Report (2022) Detailed reference viewed: 57 (3 UL)![]() Barbereau, Tom Josua ![]() ![]() ![]() in Proceedings of the Hawaii International Conference on System Sciences 2022 (2022, January) Bitcoin and Ethereum are frequently promoted as decentralized, but developers and academics question their actual decentralization. This motivates further experiments with public permissionless ... [more ▼] Bitcoin and Ethereum are frequently promoted as decentralized, but developers and academics question their actual decentralization. This motivates further experiments with public permissionless blockchains to achieve decentralization along technical, economic, and political lines. The distribution of tokenized voting rights aims for political decentralization. Tokenized voting rights achieved notoriety within the nascent field of decentralized finance (DeFi) in 2020. As an alternative to centralized crypto-asset exchanges and lending platforms (owned by companies like Coinbase and Celsius), DeFi developers typically create non-custodial projects that are not majority-owned or managed by legal entities. Holders of tokenized voting rights can instead govern DeFi projects. To scrutinize DeFi’s distributed governance strategies, we conducted a multiple-case study of non-custodial, Ethereum-based DeFi projects: Uniswap, Maker, SushiSwap, Yearn Finance, and UMA. Our findings are novel and surprising: quantitative evaluations of DeFi’s distributed governance strategies reveal a failure to achieve political decentralization. [less ▲] Detailed reference viewed: 856 (70 UL)![]() Barbereau, Tom Josua ![]() ![]() in Lacity, Mary; Treiblmaier, Horst (Eds.) Blockchains and the Token Economy: Studies in Theory and Practice (2022) Art and collectibles markets tend to involve lower liquidity and higher fees than public equity markets. Distributed ledger technology can tokenize artworks and collectibles, so that claims to these ... [more ▼] Art and collectibles markets tend to involve lower liquidity and higher fees than public equity markets. Distributed ledger technology can tokenize artworks and collectibles, so that claims to these assets can be exchanged digitally without intermediaries. Tokenization offers investors access to a global market plus a digitized paper trail, as well as new options for the fractional ownership of artworks, art-collateralized loans, and yield-bearing art assets. The main challenge for tokenization researchers and platform developers is to simultaneously satisfy regulators’ demands for transparency and auditability as well as art investors’ demands for privacy. New technological solutions are required that enable market participants to disclose the absolute minimum amount of information that is required by regulators. We explore new concepts from distributed ledger technology, cryptography, and digital identity management that can help address this challenge. [less ▲] Detailed reference viewed: 173 (20 UL)![]() Fridgen, Gilbert ![]() in Wirtschaftsinformatik & Management (2017) Detailed reference viewed: 40 (6 UL)![]() ; ; Fridgen, Gilbert ![]() in Die Bank (2012), 52(6), 46--50 Detailed reference viewed: 49 (1 UL)![]() Fridgen, Gilbert ![]() in 30th International Conference on Information Systems (ICIS) (2009) Detailed reference viewed: 36 (0 UL) |
||