arXiv subjects

Juliette Murris

Publications and source records attributed to Juliette Murris.

4 recordsLinked to original sources

arXiv2026

From Persistence to Survival: Hypothesis Testing, Effect Sizes and Vectorisation for Topological Features

Persistence diagrams are common representations in topological data analysis, but they do not naturally live in a vector space, and the statistical tools developed for comparing them have largely evolved separately from those used for downstream prediction. We introduce STRAND (Survival Topological Representation ANalysis of Diagrams), which treats (collections of) PDs as survival data: each topological feature with persistence value $p = d - b$ is a fully observed time-to-event, and the persistence survival function $S(t) = \mathbb{P}(p > t)$ is the central object for comparing diagrams. From this single representation we derive (i) a non-parametric two-sample test with calibrated Type I error and high power from a small number of diagrams; (ii) interpretable effect sizes; and (iii) a 1-Wasserstein-stable feature vector for downstream machine learning. We validate calibration and power on synthetic manifolds with controlled topology, demonstrate competitive vectorisation across 14 graph and 3D point cloud benchmarks, and apply the method to study functional brain connectivity in fMRI/neuroscience data. To our knowledge, STRAND is the first method to provide hypothesis testing and vectorisation for persistence diagrams from a single coherent and interpretable representation.

stat.ML↗

arXiv2024

Mitigating Text Toxicity with Counterfactual Generation

Toxicity mitigation consists in rephrasing text in order to remove offensive or harmful meaning. Neural natural language processing (NLP) models have been widely used to target and mitigate textual toxicity. However, existing methods fail to detoxify text while preserving the initial non-toxic meaning at the same time. In this work, we propose to apply counterfactual generation methods from the eXplainable AI (XAI) field to target and mitigate textual toxicity. In particular, we perform text detoxification by applying local feature importance and counterfactual generation methods to a toxicity classifier distinguishing between toxic and non-toxic texts. We carry out text detoxification through counterfactual generation on three datasets and compare our approach to three competitors. Automatic and human evaluations show that recently developed NLP counterfactual generators can mitigate toxicity accurately while better preserving the meaning of the initial text as compared to classical detoxification methods. Finally, we take a step back from using automated detoxification tools, and discuss how to manage the polysemous nature of toxicity and the risk of malicious use of detoxification tools. This work is the first to bridge the gap between counterfactual generation and text detoxification and paves the way towards more practical application of XAI methods.

cs.CL↗

arXiv2024

A novel methodological framework for the analysis of health trajectories and survival outcomes in heart failure patients

Heart failure (HF) contributes to circa 200,000 annual hospitalizations in France. With the increasing age of HF patients, elucidating the specific causes of inpatient mortality became a public health problematic. We introduce a novel methodological framework designed to identify prevalent health trajectories and investigate their impact on death. The initial step involves applying sequential pattern mining to characterize patients' trajectories, followed by an unsupervised clustering algorithm based on a new metric for measuring the distance between hospitalization diagnoses. Finally, a survival analysis is conducted to assess survival outcomes. The application of this framework to HF patients from a representative sample of the French population demonstrates its methodological significance in enhancing the analysis of healthcare trajectories.

stat.ME↗

arXiv2022

Towards Filling the Gaps around Recurrent Events in High-Dimensional Framework: Literature Review and Early Comparison

Background Study individuals may face repeated events overtime. However, there is no consensus around learning approaches to use in a high-dimensional framework for survival data (when the number of variables exceeds the number of individuals, i.e., p>n). This study aimed at identifying learning algorithms for analyzing/predicting recurrent events and at comparing them to standard statistical models in various data simulation settings. Methods A literature review (LR) was conducted to provide state-of-the-art methodology. Data were then simulated including variations of the number of variables and proportion of active variables. Learning algorithms from the LR were compared to standard methods in such simulation scheme. Evaluation measures were Harrell's concordance index (C-index), Kim's C-index and error rate for active variables. Results Seven publications were identified, consisting in four methodological studies, one application paper and two review. The broken adaptive ridge penalization and the RankDeepSurv deep neural network were used for comparison. On simulated data, the standard models failed when p>n. Penalized Andersen-Gill and frailty models outperformed, whereas RankDeepSurv reported lower performances. Conclusion As no guidelines support a specific approach, this study helps to better understand mechanisms and limits of investigated methods in such context.

stat.ME↗