Martin Jullum.

Research output

Publications

Browse papers and reports here, open an abstract or citation in place, or follow a link to the full text where available.

Preprints

Manuscripts under review or in preparation

  1. groupShapley: Efficient prediction explanation with Shapley values for feature groups

    Martin Jullum, Annabelle Redelmeier, Kjersti Aas

    arXiv preprint arXiv:2106.12228 · 2021

    Explainable AI#Shapley values
    DOI PDF
    Abstract

    Shapley values has established itself as one of the most appropriate and theoretically sound frameworks for explaining predictions from complex machine learning models. The popularity of Shapley values in the explanation setting is probably due to its unique theoretical properties. The main drawback with Shapley values, however, is that its computational complexity grows exponentially in the number of input features, making it unfeasible in many real world situations where there could be hundreds or thousands of features. Furthermore, with many (dependent) features, presenting/visualizing and interpreting the computed Shapley values also becomes challenging. The present paper introduces groupShapley: a conceptually simple approach for dealing with the aforementioned bottlenecks. The idea is to group the features, for example by type or dependence, and then compute and present Shapley values for these groups instead of for all individual features. Reducing hundreds or thousands of features to half a dozen or so, makes precise computations practically feasible and the presentation and knowledge extraction greatly simplified. We prove that under certain conditions, groupShapley is equivalent to summing the feature-wise Shapley values within each feature group. Moreover, we provide a simulation study exemplifying the differences when these conditions are not met. We illustrate the usability of the approach in a real world car insurance example, where groupShapley is used to provide simple and intuitive explanations.

    Cite
    APA style

    Jullum, M., Redelmeier, A., & Aas, K. (2021). groupShapley: Efficient prediction explanation with Shapley values for feature groups. arXiv preprint arXiv:2106.12228. https://doi.org/10.48550/arXiv.2106.12228

    BibTeX
    @misc{preprint-jullum2021groupshapley,
  author = {Jullum, Martin and Redelmeier, Annabelle and Aas, Kjersti},
  title = {groupShapley: Efficient prediction explanation with Shapley values for feature groups},
  year = {2021},
  howpublished = {arXiv preprint arXiv:2106.12228},
  doi = {10.48550/arXiv.2106.12228}
}

Published papers

Journal articles and conference papers

2024

  1. MCCE: Monte Carlo sampling of valid and realistic counterfactual explanations for tabular data

    Annabelle Redelmeier, Martin Jullum, Kjersti Aas, Anders Løland

    Data Mining and Knowledge Discovery · 2024

    Explainable AI#Counterfactuals
    DOI PDF
    Abstract

    We introduce MCCE: Monte Carlo sampling of valid and realistic Counterfactual Explanations for tabular data, a novel counterfactual explanation method that generates on-manifold, actionable and valid counterfactuals by modeling the joint distribution of the mutable features given the immutable features and the decision. Unlike other on-manifold methods that tend to rely on variational autoencoders and have strict prediction model and data requirements, MCCE handles any type of prediction model and categorical features with more than two levels. MCCE first models the joint distribution of the features and the decision with an autoregressive generative model where the conditionals are estimated using decision trees. Then, it samples a large set of observations from this model, and finally, it removes the samples that do not obey certain criteria. We compare MCCE with a range of state-of-the-art on-manifold counterfactual methods using four well-known data sets and show that MCCE outperforms these methods on all common performance metrics and speed. In particular, including the decision in the modeling process improves the efficiency of the method substantially.

    Cite
    APA style

    Redelmeier, A., Jullum, M., Aas, K., & Løland, A. (2024). MCCE: Monte Carlo sampling of valid and realistic counterfactual explanations for tabular data. Data Mining and Knowledge Discovery. https://doi.org/10.1007/s10618-024-01017-y

    BibTeX
    @article{redelmeier2024mcce,
  author = {Redelmeier, Annabelle and Jullum, Martin and Aas, Kjersti and Løland, Anders},
  title = {MCCE: Monte Carlo sampling of valid and realistic counterfactual explanations for tabular data},
  journal = {Data Mining and Knowledge Discovery},
  year = {2024},
  doi = {10.1007/s10618-024-01017-y}
}

2023

  1. eXplego: An interactive Tool that Helps you Select Appropriate XAI-methods for your Explainability Needs

    Martin Jullum, Jacob Sjødin, Robindra Prabhu, Anders Løland

    xAI-2023 Late-breaking Work, Demos and Doctoral Consortium Joint Proceedings · 2023

    Explainable AI#Choosing XAI methods
    PDF
    Abstract

    The growing demand for transparency, interpretability, and explainability of machine learning models and AI systems has fueled the development of methods aimed at understanding the properties and behavior of such models (XAI). Since different methods answer different explainability questions, it is crucial to understand the kind of explanation the different XAI-methods provide, and in what situations they should be used. We introduce eXplego, an interactive tree-structured tool designed to assist users in selecting the most suitable XAI method for their use case. eXplego prompts users to answer questions regarding the type of explanation they seek, guiding them along the branches of the decision tree for further inquiries. After 2-5 questions, the tree reaches one of its leaves to suggest an XAI method aligned with the user’s explainability need. The tool also provides helpful practical examples, simplified descriptions of the suggested method’s functionality and interpretability, points to consider when using the method, and links to the paper introducing the method, additional resources, and software implementations. The tool is developed from an in-depth study to discern the characteristics of the most prominent methods and the nature of the explanations they provide. We believe eXplego will help streamline the process of XAI method selection and contribute to the practical implementation of XAI in various domains. The tool is available at explego.nr.no.

    Cite
    APA style

    Jullum, M., Sjødin, J., Prabhu, R., & Løland, A. (2023). eXplego: An interactive Tool that Helps you Select Appropriate XAI-methods for your Explainability Needs. xAI-2023 Late-Breaking Work, Demos and Doctoral Consortium Joint Proceedings, 146–151.

    BibTeX
    @inproceedings{jullum2023explego,
  author = {Jullum, Martin and Sjødin, Jacob and Prabhu, Robindra and Løland, Anders},
  title = {eXplego: An interactive Tool that Helps you Select Appropriate XAI-methods for your Explainability Needs},
  booktitle = {xAI-2023 Late-breaking Work, Demos and Doctoral Consortium Joint Proceedings},
  pages = {146--151},
  year = {2023}
}

Reports & notes

Technical reports and final manuscripts

2017

  1. Statistical modeling of repertoire overlap in entire sampling spaces

    Lars Holden, Martin Jullum, Geir Kjetil Sandve

    NR-note SAMBA/13/17, Norwegian Computing Center · 2017

    Statistical modellingMachine learning in practice
    Abstract

    We analyze the distribution of T-cell clonotypes in a compartment like blood based on samples. In particular, we study how the distribution of clonotype frequencies changes between different samples. We consider this as a sampling problem and formulate the problem as a generalization of the classical statistical problem of comparing samples from an urn. Due to the low sampling size compared to the number of different clonotypes in the entire sampling space, the classical methodology that works directly with clonotype frequencies in samples is not suited. We approach this challenge by representing other properties of the sample. Our re-representation allows for easy sampling model fitting and testing under natural model conditions. Although we here focus on the application on clonotypes, the new methodology generalizes seamlessly to other applications.

    Cite
    APA style

    Holden, L., Jullum, M., & Sandve, G. K. (2017). Statistical modeling of repertoire overlap in entire sampling spaces (NR Note SAMBA/13/17). Norwegian Computing Center.

    BibTeX
    @techreport{holden2017repertoire,
  author = {Holden, Lars and Jullum, Martin and Sandve, Geir Kjetil},
  title = {Statistical modeling of repertoire overlap in entire sampling spaces},
  institution = {Norwegian Computing Center},
  type = {NR note},
  number = {SAMBA/13/17},
  year = {2017}
}