Heterogeneity across sub-populations can be beneficially exploited for causal inference and to improve the replicability of feature selections across distributions. The key is to encourage models to exhibit invariance across settings and interventions. The novel methodology potentially offers more robustness and 'causal-oriented' interpretation of results, compared to standard regression and classification methods.
● "Invariance, Causality and Robustness," https://arxiv.org/pdf/1812.08233
● "Anchor regression: heterogeneous data meets causality,"
● "Causal Dantzig: Fast inference in linear structural equation models with hidden
variables under additive interventions,"
The Workshop is held from 1:30-2:50pm in Medical School Office Building (MSOB), Rm x303, 1265 Welch Road, Stanford, unless otherwise specified on the calendar at the link below.