mlsanityData to 7 Oct 2026

Attention-based Dynamic Multilayer Graph Neural Networks for Loan Default Prediction

arXiv 2402.00299 ↗First posted 1 Feb 2024q-fin.GN cs.LGcited by 5

Not on the map. Why?

Abstract (arXiv)

Whereas traditional credit scoring tends to employ only individual borrower- or loan-level predictors, it has been acknowledged for some time that connections between borrowers may result in default risk propagating over a network. In this paper, we present a model for credit risk assessment leveraging a dynamic multilayer network built from a Graph Neural Network and a Recurrent Neural Network, each layer reflecting a different source of network connection. We test our methodology in a behavioural credit scoring context using a dataset provided by U.S. mortgage financier Freddie Mac, in which different types of connections arise from the geographical location of the borrower and their choice of mortgage provider. The proposed model considers both types of connections and the evolution of these connections over time. We enhance the model by using a custom attention mechanism that weights the different time snapshots according to their importance. After testing multiple configurations, a model with GAT, LSTM, and the attention mechanism provides the best results. Empirical results demonstrate that, when it comes to predicting probability of default for the borrowers, our proposed model brings both better results and novel insights for the analysis of the importance of connections and timestamps, compared to traditional methods.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Builds on 41 of its 81 references that are in the map. First posted on arXiv 1 Feb 2024.

Builds on

41 of 81 references.

References · 41

  1. 2017Attention Is All You Needcited by 59,932
  2. 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
  3. 2017Graph Attention Networkscited by 8,854
  4. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  5. 2014Neural Machine Translation by Jointly Learning to Align and Translatecited by 4,530
  6. 2014Sequence to Sequence Learning with Neural Networkscited by 2,974
  7. 2014Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modelingcited by 2,341
  8. 2018Graph Neural Networks: A Review of Methods and Applicationscited by 1,571
  9. 2015Show, Attend and Tell: Neural Image Caption Generation with Visual Attentioncited by 1,421
  10. 2014On the Properties of Neural Machine Translation: Encoder-Decoder Approachescited by 1,303
  11. 2017Spatio-Temporal Graph Convolutional Networks: A Deep Learning Framework for Traffic Forecastingcited by 1,125
  12. 2015Effective Approaches to Attention-based Neural Machine Translationcited by 1,024
29 more
  1. 2018Link Prediction Based on Graph Neural Networkscited by 829
  2. 2021Deep Neural Networks and Tabular Data: A Surveycited by 489
  3. 2014Long-term Recurrent Convolutional Networks for Visual Recognition and Descriptioncited by 483
  4. 2020Inductive Representation Learning on Temporal Graphscited by 454
  5. 2019EvolveGCN: Evolving Graph Convolutional Networks for Dynamic Graphscited by 443
  6. 2019Predicting Dynamic Embedding Trajectory in Temporal Interaction Networkscited by 443
  7. 2018T-GCN: A Temporal Graph ConvolutionalNetwork for Traffic Predictioncited by 399
  8. 2017A Structured Self-attentive Sentence Embeddingcited by 318
  9. 2019Representation Learning for Dynamic Graphs: A Surveycited by 260
  10. 2015Structural-RNN: Deep Learning on Spatio-Temporal Graphscited by 234
  11. 2020Graph Clustering with Graph Neural Networkscited by 163
  12. 2016Long Short-Term Memory-Networks for Machine Readingcited by 161
  13. 2017Attentional Factorization Machines: Learning the Weight of Feature Interactions via Attention Networkscited by 126
  14. 2020Foundations and modelling of dynamic networks using Dynamic Graph Neural Networks: A surveycited by 123
  15. 2013Multilayer Networkscited by 86
  16. 2017Dynamic Graph Convolutional Networkscited by 83
  17. 2018An LSTM Network for Highway Trajectory Predictioncited by 81
  18. 2021A Review on Graph Neural Network Methods in Financial Applicationscited by 77
  19. 2021Graph Learning based Recommender Systems: A Reviewcited by 59
  20. 2015Deep Sentence Embedding Using Long Short-Term Memory Networks: Analysis and Application to Information Retrievalcited by 54
  21. 2018Dynamic Graph Representation Learning via Self-Attention Networkscited by 39
  22. 2020The Value of Big Data for Credit Scoring: Enhancing Financial Inclusion using Mobile Phone Data and Social Network Analyticscited by 21
  23. 2017Multilingual Hierarchical Attention Networks for Document Classificationcited by 9
  24. 2020Super-App Behavioral Patterns in Credit Risk Models: Financial, Statistical and Regulatory Implicationscited by 9
  25. 2018GrAMME: Semi-Supervised Learning using Multi-layered Graph Attention Modelscited by 8
  26. 2020Multilayer Network Analysis for Improved Credit Risk Predictioncited by 8
  27. 2020Evolution of Credit Risk Using a Personalized Pagerank Algorithm for Multilayer Networkscited by 4
  28. 2021mGNN: Generalizing the Graph Neural Networks to the Multilayer Casecited by 4
  29. 2018Deep Canonically Correlated LSTMscited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 3; 2026: 2.

5 papers.

  1. 2026Foundation Models for Credit Risk Prediction: A Game Changer?cited by 2
  2. 2025Kolmogorov-Arnold Networks-based GRU and LSTM for Loan Default Early Predictioncited by 0
  3. 2025Transforming Credit Risk Analysis: A Time-Series-Driven ResE-BiLSTM Framework for Post-Loan Default Detectioncited by 0
  4. 2025A Multimodal Approach to SME Credit Scoring Integrating Transaction and Ownership Networkscited by 0
  5. 2026Communicating Credit Risk with Large Language Models: Evaluation of Explanations from Standard and Alternative Data-Based Modelscited by 0