arXiv:2607. 18883v1 Announce Type: cross Abstract: A central aim of unsupervised learning is to uncover latent factors that explain dependencies among observations.
We develop a general method for amortized Bayesian inference on multilevel models of arbitrary structure. Given a generative model specified as a directed acyclic graph, our method automatically deriv...
arXiv:2609.40024v1 Announce Type: new Abstract: We develop a general method for amortized Bayesian inference on multilevel models of arbitrary structure. Given a generative model specified as a direc...
arXiv:2606. 17516v1 Announce Type: cross Abstract: Causal discovery from observational data remains challenging due to the need to recover directed structure and latent confounding without interventions.
arXiv:2606. 15458v1 Announce Type: cross Abstract: Variational inference (VI) is a core engine of modern AI, enabling scalable approximate Bayesian learning and uncertainty-aware training of large probabilistic and generative models.
arXiv:2603. 19139v2 Announce Type: replace Abstract: Learning systems must balance generalization across experiences with discrimination of task-relevant details.
The paper introduces a variational inference framework that jointly discovers latent clusters of variables and the causal relationships among those clusters. It models clusters with categorical distributions and graph structures with Bernoulli distributions, deriving variational lower bounds and estimation techniques for learning both cluster assignments and causal links. The method’s effectiveness is shown on synthetic and real datasets.
arXiv:2605. 03517v4 Announce Type: replace Abstract: Self-supervised learning (SSL) excels at finding general-purpose latent representations from complex data, yet lacks a unifying theoretical framework that explains the diverse existing methods and guides the design of new ones.
Neuro-Causal Factor Analysis (NCFA) reimagines traditional factor analysis by integrating causal structure learning and deep generative modeling. The method learns a directed graph linking latent and observed variables, then trains a deep generative model that respects the graph’s Markov factorization. Experiments on synthetic and real datasets show NCFA achieves lower reconstruction error than standard FA and better latent distribution recovery than a variational autoencoder, while offering a sparser architecture, reduced complexity, and causal interpretability.
arXiv:2607. 23337v1 Announce Type: new Abstract: Neural operators provide data-driven mappings for modeling dynamical systems.
arXiv:2606. 19366v1 Announce Type: cross Abstract: Information lattice learning (ILL) learns interpretable rules of a signal by alternately projecting the signal onto a partition lattice that encodes a hierarchy of abstractions and lifting selected rules back to the signal domain.
arXiv:2605. 15995v2 Announce Type: replace-cross Abstract: Learning latent representations from complex data is central to modern machine learning, spanning temporal, multimodal, and partially observed systems.
arXiv:2603.20111v2 Announce Type: replace Abstract: The Joint-Embedding Predictive Architecture (JEPA) is often seen as a non-generative alternative to likelihood-based self-supervised learning, emph...