/pdf/counterfactuals-and-causability-in-explainable-artificial-15jjx262.pdf

Counterfactuals and causability in explainable artificial intelligence: Theory, algorithms, and applications

The level of modeling sophistication in financial services has increased considerably over the years Nowadays, the complexity of financial problems and the vast amount of data require an engineering approach based on analytical modeling tools for planning, decision making, reporting, and supervisory control This article provides an overview of the main financial applications of computational and data analytics approaches, focusing on the coverage of the recent developments and trends The overview covers different methodological tools and their uses in areas, such as portfolio management, credit analysis, banking, and insurance (This abstract was borrowed from another version of this item)

Computational approaches and data analytics in financial services: A literature review

The use of sophisticated machine learning models for critical decision-making faces the challenge that these models are often applied as a ‘black-box’. This has led to an increased interest in interpretable machine learning, where post-hoc model-agnostic algorithms present a useful mechanism for generating interpretations of complex learning models. This paper proposes a novel approach based on Bayesian Networks to generate local post-hoc model-agnostic interpretations of a black-box predictive model. Consequently, the proposed approach presents features that are conditionally dependent between each other and that are directly influencing the class variable. This enables the decision-maker to better understand how features are related and why a certain prediction was made. Compared to the existing post-hoc interpretation methods, the contribution of our approach is three-fold: (1) as a probabilistic graphical model, the extracted Bayesian network can provide interpretations through conditional dependencies in a graphical structure regarding what input features and how/why they contributed to a prediction; (2) for complex decision problems with many features, a Markov blanket can be generated from the extracted Bayesian network to provide interpretations with a focused view on those input features that directly contributed to a prediction; (3) the extracted Bayesian network enables the identification of four different rules which can inform the decision-maker about the confidence level in a prediction, thus helping the decision-maker assess the reliability of predictions learned by a black-box model. We implemented the proposed approach, applied it in the context of two well-known public datasets and analysed the results, which are made available in an open-source repository: https://github.com/catarina-moreira/LINDA_DSS .

LINDA-BN: An interpretable probabilistic approach for demystifying black-box predictive models

In credit risk estimation, the most important element is obtaining a probability of default as close as possible to the effective risk. This effort quickly prompted new, powerful algorithms that reach a far higher accuracy, but at the cost of losing intelligibility, such as Gradient Boosting or ensemble methods. These models are usually referred to as "black-boxes", implying that you know the inputs and the output, but there is little way to understand what is going on under the hood. As a response to that, we have seen several different Explainable AI models flourish in recent years, with the aim of letting the user see why the black-box gave a certain output. In this context, we evaluate two very popular eXplainable AI (XAI) models in their ability to discriminate observations into groups, through the application of both unsupervised and predictive modeling to the weights these XAI models assign to features locally. The evaluation is carried out on real Small and Medium Enterprises data, obtained from official italian repositories, and may form the basis for the employment of such XAI models for post-processing features extraction.

SHAP and LIME: An Evaluation of Discriminative Power in Credit Risk.

Peer-to-peer (P2P) lending demands effective and explainable credit risk models. Typical machine learning algorithms offer high prediction performance, but most of them lack explanatory power. However, this deficiency can be solved with the help of the explainability tools proposed in the last few years, such as the SHAP values. In this work, we assess the well-known logistic regression model and several machine learning algorithms for granting scoring in P2P lending. The comparison reveals that the machine learning alternative is superior in terms of not only classification performance but also explainability. More precisely, the SHAP values reveal that machine learning algorithms can reflect dispersion, nonlinearity and structural breaks in the relationships between each feature and the target variable. Our results demonstrate that is possible to have machine learning credit scoring models be both accurate and transparent. Such models provide the trust that the industry, regulators and end-users demand in P2P lending and may lead to a wider adoption of machine learning in this and other risk assessment applications where explainability is required.

Miller Janny Ariza-Garzon

Papers

Explainability of a Machine Learning Granting Scoring Model in Peer-to-Peer Lending

Risk-Return modelling in the P2P lending market: Trends, Gaps, Recommendations and future directions