Modeling financial distress propagation on customer-supplier networks.

Financial networks have been the object of intense quantitative analysis during the last few decades. Their structure and the dynamical processes on top of them are of utmost importance to understand the emergent collective behavior behind economic and financial crises. In this paper, we propose a stylized model to understand the "domino effect" of distress in client-supplier networks. We provide a theoretical analysis of the model, and we apply it to several synthetic networks and a real customer-supplier network, supplied by one of the largest banks in Europe. Besides, the proposed model allows us to investigate possible scenarios for the functioning of the financial distress propagation and to assess the economic health of the full network. The main novelty of this model is the combination of two stochastic terms: an additive noise, accounting by the capability of trading and paying obligations, and a multiplicative noise representing the variations of the market. Both parameters are crucial to determining the maximum default probability and the diffusion process characteristics.

Differential Replication for Credit Scoring in Regulated Environments.

Differential replication is a method to adapt existing machine learning solutions to the demands of highly regulated environments by reusing knowledge from one generation to the next. Copying is a technique that allows differential replication by projecting a given classifier onto a new hypothesis space, in circumstances where access to both the original solution and its training data is limited. The resulting model replicates the original decision behavior while displaying new features and characteristics. In this paper, we apply this approach to a use case in the context of credit scoring. We use a private residential mortgage default dataset. We show that differential replication through copying can be exploited to adapt a given solution to the changing demands of a constrained environment such as that of the financial market. In particular, we show how copying can be used to replicate the decision behavior not only of a model, but also of a full pipeline. As a result, we can ensure the decomposability of the attributes used to provide explanations for credit scoring models and reduce the time-to-market delivery of these solutions.

Environmental Adaptation and Differential Replication in Machine Learning.

When deployed in the wild, machine learning models are usually confronted with an environment that imposes severe constraints. As this environment evolves, so do these constraints. As a result, the feasible set of solutions for the considered need is prone to change in time. We refer to this problem as that of environmental adaptation. In this paper, we formalize environmental adaptation and discuss how it differs from other problems in the literature. We propose solutions based on differential replication, a technique where the knowledge acquired by the deployed models is reused in specific ways to train more suitable future generations. We discuss different mechanisms to implement differential replications in practice, depending on the considered level of knowledge. Finally, we present seven examples where the problem of environmental adaptation can be solved through differential replication in real-life applications.

Risk mitigation in algorithmic accountability: The role of machine learning copies.

Machine learning plays an increasingly important role in our society and economy and is already having an impact on our daily life in many different ways. From several perspectives, machine learning is seen as the new engine of productivity and economic growth. It can increase the business efficiency and improve any decision-making process, and of course, spawn the creation of new products and services by using complex machine learning algorithms. In this scenario, the lack of actionable accountability-related guidance is potentially the single most important challenge facing the machine learning community. Machine learning systems are often composed of many parts and ingredients, mixing third party components or software-as-a-service APIs, among others. In this paper we study the role of copies for risk mitigation in such machine learning systems. Formally, a copy can be regarded as an approximated projection operator of a model into a target model hypothesis set. Under the conceptual framework of actionable accountability, we explore the use of copies as a viable alternative in circumstances where models cannot be re-trained, nor enhanced by means of a wrapper. We use a real residential mortgage default dataset as a use case to illustrate the feasibility of this approach.

The keys to control a COVID-19 outbreak in a haemodialysis unit.

BACKGROUND: The high rate of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) spreading represents a challenge to haemodialysis (HD) units. While fast isolation of suspected cases plays an essential role to avoid disease outbreaks, significant rates of asymptomatic cases have recently been described. After detecting an outbreak in one of our HD clinics, wide SARS-CoV-2 screening and segregation of confirmed cases were performed. METHODS: The entire clinic population, 192 patients, underwent testing for SARS-CoV-2 detection by real-time reverse-transcriptase polymerase chain reaction . We used univariate and multivariate logistic regression to define variables involved in SARS-CoV-2 infection in our dialysis unit. Later, we analysed differences between symptomatic and asymptomatic SARS-CoV-2-positive patients. RESULTS: In total, 22 symptomatic and 14 of the 170 asymptomatic patients had a SARS-CoV-2-positive result. Living in a nursing home/homeless [odds ratio (OR) 3.54; P = 0.026], having been admitted to the reference hospital within the previous 2 weeks (OR 5.19; P = 0.002) and sharing health-care transportation with future symptomatic (OR 3.33; P = 0.013) and asymptomatic (OR 4.73; P = 0.002) positive patients were independent risk factors for a positive test. Nine positive patients (25.7%) remained asymptomatic after a 3-week follow-up. We found no significant differences between symptomatic and asymptomatic SARS-CoV-2-positive patients. CONCLUSIONS: Detection of asymptomatic SARS-CoV-2-positive patients is probably one of the key points to controlling an outbreak in an HD unit. Sharing health-care transportation to the dialysis unit, living in a nursing home and having been admitted to the reference hospital within the previous 2 weeks, are major risk factors for SARS-CoV-2 infection.