5 papers
Interventional Processes for Causal Uncertainty Quantification
Hugh Dance, Peter Orbanz, Arthur Gretton
Reliable uncertainty quantification for causal effects is crucial in high-stakes applications, but remains challenging when the target is an entire function rather than a scalar es…
Debiased Counterfactual Generation via Flow Matching from Observations
Hugh Dance, Johnny Xi, Peter Orbanz +1
Estimating counterfactual distributions under interventions is central to treatment risk assessment and counterfactual generation tasks. Existing approaches model the counterfactua…
Counterfactual Cocycles: A Framework for Robust and Coherent Counterfactual Transports
Hugh Dance, Benjamin Bloem-Reddy
Estimating joint distributions (a.k.a. couplings) over counterfactual outcomes is central to personalized decision-making and treatment risk assessment. Two emergent frameworks wit…
Efficiently Vectorized MCMC on Modern Accelerators
Hugh Dance, Pierre Glaser, Peter Orbanz +1
With the advent of automatic vectorization tools (e.g., JAX's ), writing multi-chain MCMC algorithms is often now as simple as invoking those tools on single-chain c…
Distinguishing Cause from Effect with Causal Velocity Models
Johnny Xi, Hugh Dance, Peter Orbanz +1
Bivariate structural causal models (SCM) are often used to infer causal direction by examining their goodness-of-fit under restricted model classes. In this paper, we describe a pa…