Machine Learning (stat.ML)

Abstract: The arrival of digital platforms has revolutionized occupational health by giving the possibility to Occupational Health Services (SPSTI) to acquire databases to offer professionals new possibilities for action. However, in a sector of activity that has been questioning the development of multidisciplinarity for 20 years, the arrival of new tools can sometimes seem to be a quick solution. The study, conducted in a precursor SPSTI in terms of the development of digital tools, aims to take stock of the methods and impacts of instrumental and organizational transformations for health professionals as well as for members of the technical teams of the SPSTI. It is a question of highlighting the brakes and the levers as well as the various possibilities of accompaniment to consider.

Abstract: We propose a novel approach for interpreting deep embeddings in the context of patient clustering. We evaluate our approach on a dataset of participants with type 2 diabetes from the UK Biobank, and demonstrate clinically meaningful insights into disease progression patterns.

Abstract: Causal disentanglement aims to uncover a representation of data using latent variables that are interrelated through a causal model. Such a representation is identifiable if the latent model that explains the data is unique. In this paper, we focus on the scenario where unpaired observational and interventional data are available, with each intervention changing the mechanism of a latent variable. When the causal variables are fully observed, statistically consistent algorithms have been developed to identify the causal model under faithfulness assumptions. We here show that identifiability can still be achieved with unobserved causal variables, given a generalized notion of faithfulness. Our results guarantee that we can recover the latent causal model up to an equivalence class and predict the effect of unseen combinations of interventions, in the limit of infinite data. We implement our causal disentanglement framework by developing an autoencoding variational Bayes algorithm and apply it to the problem of predicting combinatorial perturbation effects in genomics.