Institut de Recherche Technologique SystemX

Non-asymptotic oracle inequalities for the Lasso in high-dimensional mixture of experts

Abstract We study Lasso estimation for high-dimensional softmax-gated Gaussian mixture-of-experts (SGMoE) regression models, where the predictor dimension P may exceed the sample size N . Under bounded parameters, we establish a non-asymptotic Kullback–Leibler (KL) oracle inequal...

Recovery trajectories and predictors of symptom resolution in post-COVID-19 condition: a population-based cohort study

Background SARS-CoV-2 infection can lead to persistent symptoms, known as Post-COVID-19 Condition (PCC). Previous studies on PCC prognosis mainly looked at severe cases in rehabilitation settings and without knowledge about pre-infection symptom levels. Uncertainty remains about...