{"ID":2898423,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.03622","arxiv_id":"2507.03622","title":"Localising Dropout Variance in Twin Networks","abstract":"Accurate individual treatment-effect estimation demands not only reliable point predictions but also uncertainty measures that help practitioners \\emph{locate} the source of model failure. We introduce a layer-wise variance decomposition for deep twin-network models: by toggling Monte Carlo Dropout independently in the shared encoder and the outcome heads, we split total predictive variance into an \\emph{encoder component} ($σ_{\\mathrm{enc}}^2$) and a \\emph{head component} ($σ_{\\mathrm{head}}^2$), with $σ_{\\mathrm{enc}}^2 + σ_{\\mathrm{head}}^2 \\approx σ_{\\mathrm{tot}}^2$ by the law of total variance. Across three synthetic covariate-shift regimes, the encoder component dominates under distributional shift ($ρ_{\\mathrm{enc}}=0.53$) while the head component becomes informative only once encoder uncertainty is controlled. On a real-world twins cohort with induced multivariate shift, only $σ_{\\mathrm{enc}}^2$ spikes on out-of-distribution samples and becomes the primary error predictor ($ρ_{\\mathrm{enc}}\\!\\approx\\!0.89$), while $σ_{\\mathrm{head}}^2$ remains flat. The decomposition adds negligible cost over standard MC Dropout and provides a practical diagnostic for deciding whether to collect more diverse covariates or more outcome data.","short_abstract":"Accurate individual treatment-effect estimation demands not only reliable point predictions but also uncertainty measures that help practitioners \\emph{locate} the source of model failure. We introduce a layer-wise variance decomposition for deep twin-network models: by toggling Monte Carlo Dropout independently in the...","url_abs":"https://arxiv.org/abs/2507.03622","url_pdf":"https://arxiv.org/pdf/2507.03622v2","authors":"[\"Cooper Doyle\"]","published":"2025-07-04T14:48:51Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.AI\",\"stat.ML\"]","methods":"[]","has_code":false}