analiza aiqbench
This arXiv preprint proposes CauVaDE, a causal generative model that represents an unobserved confounder as a discrete latent cluster, and it argues that interventional image distributions are generally non-identifiable from observational data. For readers, the practical point is that a single generative model can fit the training data well yet support many different causal explanations, so claims about controllable attributes such as age or sex depend on modelling assumptions. The paper's theoretical approximation results are the main contribution, but its empirical support for the diversity-generating bootstrap will determine how useful it is.
A Causal Bootstrap for Diverse Image Generation under Unobserved Confounding
arXiv:2606.21806v3 Announce Type: replace Abstract: Deep generative models reproduce the observational distribution underlying the training data, which, in turn, inherit any spurious associations it contains. A common source is an unobserved confounder that simultaneously affects both a treatment attribute the user wants to control/edit (e.g., age, sex) and an observed outcome (e.g., an image) expected to vary in response. Existing causal inference methods resolve the resulting ambiguity by imposing structural assumptions strong enough to single out one interventional distribution. However, in practical domains like image generation, such assumptions are rarely warranted, and the interventional distribution is generally non-identifiable: distinct causal models can agree on the observed data yet imply different causal predictions. This paper proposes CauVaDE (Causal Variational Deep Embedding), a novel family of causal generative models in which the unobserved confounder collapses into a discrete latent cluster with bounded support without loss of generality. We prove that this family can approximate both observational and interventional image distributions with arbitrary accuracy in a canonical latent causal diagram. Building on this model representation, we introduce a novel causal bootstrap that varies the strength of confounder entropy to generate a diverse family of causal explanations, each defining a distinct interventional image distribution. Such a procedure produces distinct interventional images while maintaining a good fit to the observed data.
Abstract:Deep generative models reproduce the observational distribution underlying the training data, which, in turn, inherit any spurious associations it contains. A common source is an unobserved confounder that simultaneously affects both a treatment attribute the user wants to control/edit (e.g., age, sex) and an observed outcome (e.g., an image) expected to vary in response. Existing causal inference methods resolve the resulting ambiguity by imposing structural assumptions strong enough to single out one interventional distribution. However, in practical domains like image generation, such assumptions are rarely warranted, and the interventional distribution is generally non-identifiable: distinct causal models can agree on the observed data yet imply different causal predictions. This paper proposes CauVaDE (Causal Variational Deep Embedding), a novel family of causal generative models in which the unobserved confounder collapses into a discrete latent cluster with bounded support without loss of generality. We prove that this family can approximate both observational and interventional image distributions with arbitrary accuracy in a canonical latent causal diagram. Building on this model representation, we introduce a novel causal bootstrap that varies the strength of confounder entropy to generate a diverse family of causal explanations, each defining a distinct interventional image distribution. Such a procedure produces distinct interventional images while maintaining a good fit to the observed data.
Machine Learning (cs.LG)
arXiv:2606.21806 [cs.LG]
(or arXiv:2606.21806v3 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2606.21806 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 303 cuvinte
Subiecte conexe
Identifying Causal Effects Using a Single Proxy Variable
arXiv · 28 sept. 2026
Point-Identification of a Robust Predictor Under Latent Shift with Imperfect Proxies
arXiv · 12 iun. 2026
Point-Identification of a Robust Predictor Under Latent Shift with Imperfect Proxies
arXiv · 11 iun. 2026
Concept Drift from a Causal Perspective
arXiv · 23 sept. 2026