conditional distribution modeling

This study addresses the fundamental statistical problem of conditional density estimation by systematically comparing classical nonparametric approaches—such as single-index models, basis expansion methods (e.g., FlexCode), and DeepCDE—with modern generative models, including conditional GANs and conditional denoising diffusion probabilistic models. For the first time, these methods are evaluated within a unified and reproducible framework using metrics like mean squared error and Wasserstein distance to assess their accuracy, flexibility, and computational cost in estimating conditional means and standard deviations. The analysis clarifies the performance boundaries and practical applicability of each approach, offering both theoretical insights and actionable guidance for selecting appropriate methods in predictive modeling, uncertainty quantification, and probabilistic inference tasks.

This work addresses the longstanding limitation in conditional density estimation—namely, the absence of closed-form solutions for multivariate conditional densities under non-Gaussian assumptions. We propose a generative conditional density estimation framework grounded in copula modeling and analytic conditionalization in latent space. Methodologically, we first establish the inheritability of “conditional stability” under mixture and transformation operations, thereby extending analytically tractable conditional families to non-Gaussian, nonlinear, and cross-dimensional settings. The core components include a Gaussian Mixture Copula Model (GMCM), an explicit latent-space conditionalization mechanism, and joint copula modeling. Experiments on synthetic and real-world datasets demonstrate substantial improvements in conditional density estimation accuracy and robustness to missing data imputation. Crucially, our approach enables efficient, differentiable, and sampling-free deterministic conditional inference.

This work addresses the challenge of zero-shot conditional generation from pretrained unconditional diffusion models—specifically, generating samples satisfying complex logical constraints (e.g., structural conditions on tables, images, or time series) without fine-tuning. We propose a neural-symbolic soft-constraint embedding method that encodes first-order logic constraints as differentiable soft penalties and directly perturbs the score function to achieve theoretically consistent approximation of the conditional distribution—bypassing classifier-guided sampling or costly retraining. Our approach integrates score-based modeling, symbolic logic encoding, score correction, and stabilized sampling. Experiments across diverse data modalities demonstrate that our method achieves high-fidelity approximation of the true conditional distribution, significantly outperforming existing zero-shot conditional generation baselines.

This work addresses weak interpolation controllability and training instability in conditional generation. We propose Conditional Stochastic Interpolation (CSI), a framework that enables differentiable and controllable transport from a reference distribution to a target conditional distribution by modeling conditional probability flows or stochastic differential equations (SDEs). Key contributions include: (i) the first explicit formulation of the conditional drift and score function as conditional expectations; (ii) an adaptive diffusion term that enhances training stability; (iii) a non-asymptotic error bound guaranteeing convergence and generalization; and (iv) support for parameter-free regression estimation, deterministic ODE sampling, and adaptive-diffusion sampling. Extensive experiments on standard image datasets demonstrate high-quality, high-fidelity conditional generation. The method combines theoretical rigor—grounded in conditional probability flow theory—with practical effectiveness, offering improved controllability, stability, and flexibility over existing approaches.

To address the limited discriminative capability of naïve Bayes stemming from its strong conditional independence (isotropic) assumption, this paper proposes Projection Naïve Bayes (PNB), which learns an optimal linear subspace via discriminative projection optimization and performs naïve Bayes factorization of class-conditional densities within this low-dimensional projected space. PNB is the first framework to deeply integrate discriminative projection learning with naïve Bayes modeling, simultaneously enabling dimensionality reduction, visualization, and theoretical interpretability; it is further shown to be equivalent to class-conditional independent component analysis. Extensive experiments across 162 public benchmark datasets demonstrate that PNB significantly outperforms classical probabilistic discriminative models—including Linear Discriminant Analysis (LDA) and Quadratic Discriminant Analysis (QDA)—and matches the accuracy of Support Vector Machines (SVM), while retaining the statistical interpretability and computational efficiency inherent to generative models.

This work proposes a nonparametric inference method for conditional functionals—such as the conditional mean—in settings where labeled data are scarce, unlabeled covariates are abundant, and a black-box predictor is available. The approach avoids parametric modeling assumptions by leveraging a data-adaptive kernel localization and a prediction-correction decomposition, which transforms conditional moment estimation into a weighted unconditional moment problem while incorporating the black-box predictor to reduce variance. Theoretical analysis establishes non-asymptotic error bounds, minimax optimal convergence rates, and asymptotic normality, along with an explicit variance decomposition that quantifies the contributions of both the predictor and the unlabeled data. Experiments demonstrate that the resulting confidence intervals achieve accurate coverage and are significantly narrower, marking the first method in a fully nonparametric framework to simultaneously attain validity and efficiency gains.

This work addresses the limitations of deterministic regression in settings involving coarse-graining, partial observability, or inverse problems, where input–output relationships are inherently one-to-many and conditional distributions exhibit irreducible stochasticity. To diagnose these challenges under finite data, the authors introduce a framework centered on the “conditional mean barrier,” proposing two diagnostic tools: a residual–feature orthogonality test and an upper-bound analysis of the coefficient of determination. These tools effectively disentangle model underfitting from irreducible conditional variance. Leveraging this diagnostic framework, the study systematically evaluates distribution-learning approaches—including negative log-likelihood, moment matching, variational objectives, adversarial divergences, and score matching—on benchmark problems such as bimodal distributions and multiscale Lorenz-96 closure tasks. Empirical results demonstrate that the framework clearly identifies the inadequacies of deterministic models and reveals the true variability of underlying conditional distributions.

This work addresses the limitation of traditional methods that learn conditional distributions separately for fixed joint distributions, thereby struggling to generalize to unseen distribution pairs. It reframes conditional probability modeling as a universal operator learning problem across distributions and proposes a single neural operator that amortizes the approximation of conditional densities by mapping arbitrary joint densities to their corresponding conditionals. Leveraging neural operator theory and continuity analysis in spaces of probability densities, the authors establish that this operator is continuous over suitable classes of densities and can be arbitrarily well approximated by neural networks. Empirical validation on Gaussian mixture distributions demonstrates the framework’s effectiveness, showcasing strong generalization capabilities and promising applicability in tasks such as Bayesian inference.

Traditional inverse design methods are limited to point-wise target outputs and struggle to accommodate design requirements expressed as target distributions. This work formalizes, for the first time, the distribution-level inverse design problem and introduces a new paradigm termed Conditional Distribution Matching (CDM), defining two task variants: CDMS and CDMO. The authors propose MLGD-F, a plug-and-play inference algorithm that efficiently solves these tasks without additional training. MLGD-F leverages a pre-trained score-based diffusion model combined with a single-step conditional sampler, using a matching loss to guide gradient updates. The method successfully recovers inputs whose outputs align with complex target distributions—including discrete mixtures and continuous low-rank supports—demonstrating effectiveness across synthetic data, structured image transformation, and generative editing tasks.

This work proposes the Ens-CGP framework to establish a unified probabilistic representation for ensemble data, bridging the theoretical gap among ensemble methods, variational inference, and Gaussian processes. By treating empirical ensemble moments as (possibly low-rank) Gaussian priors, the approach constructs a conditional Gaussian process (CGP) through exact Bayesian conditioning. The framework explicitly separates representation—progressing from GP to CGP to Ens-CGP—from computational algorithms such as the ensemble Kalman filter (EnKF), thereby revealing the common probabilistic foundation underlying Kalman filtering, maximum a posteriori (MAP) estimation, and RKHS regularized regression under conditional Gaussian laws. This study provides a rigorous probabilistic basis for ensemble-based inference and, for the first time, unifies the probabilistic, variational, and ensemble perspectives at both geometric and representational levels.