Thus, one of the most fundamental tasks in statistics is to provide bounds on the sample size which is sufficient to soundly represent the population, and probabilistic tools are used to derive such guarantees, under a variety of assumptions. Virtually all of these guarantees are based on classical probabilistic models which assume that the target population is fixed in advance and does not depend on the sample collected throughout the process . Such an assumption, that the setting is offline (or oblivious or static), is however not always realistic. In this work we explore an abstract framework which removes this assumption, and prove that natural and efficient sampling processes produce samples which soundly represent the target population. Situations where the sampling process explicitly or implicitly affects the target population are abundant in modern data analysis. Consider, for instance, navigation apps that optimize traffic by routing drivers to less congested routes: such apps collect statistics from drivers to estimate the traffic-load on the routes, and use these estimates to guide their users through faster routes.

We consider the problem of representation and learning of causal models that encode context-specific information for discrete data. To represent such models we define the class of CStrees. This class is a subclass of staged tree models that captures context-specific information in a DAG model by the use of a staged tree, or equivalently, by a collection of DAGs. We provide a characterization of the complete set of asymmetric conditional independence relations encoded by a CStree that generalizes the global Markov property for DAGs. As a consequence, we obtain a graphical characterization of model equivalence for CStrees generalizing that of Verma and Pearl for DAG models. We also provide a closed-form formula for the maximum likelihood estimator of a CStree and use it to show that the Bayesian Information Criterion is a locally consistent score function for this model class. We then use the theory for general interventions in staged tree models to provide a global Markov property and a characterization of model equivalence for general interventions in CStrees. As examples, we apply these results to two real data sets, learning BIC-optimal CStrees for each and analyzing their context-specific causal structure.

Score-based generative modeling has recently emerged as a promising alternative to traditional likelihood-based or implicit approaches. Learning in score-based models involves first perturbing data with a continuous-time stochastic process, and then matching the time-dependent gradient of the logarithm of the noisy data density - or score function - using a continuous mixture of score matching losses. In this note, we show that such an objective is equivalent to maximum likelihood for certain choices of mixture weighting. This connection provides a principled way to weight the objective function, and justifies its use for comparing different score-based generative models. Taken together with previous work, our result reveals that both maximum likelihood training and test-time log-likelihood evaluation can be achieved through parameterization of the score function alone, without the need to explicitly parameterize a density function.

Stacking is a widely used model averaging technique that yields asymptotically optimal prediction among all linear averages. We show that stacking is most effective when the model predictive performance is heterogeneous in inputs, so that we can further improve the stacked mixture with a hierarchical model. With the input-varying yet partially-pooled model weights, hierarchical stacking improves average and conditional predictions. Our Bayesian formulation includes constant-weight (complete-pooling) stacking as a special case. We generalize to incorporate discrete and continuous inputs, other structured priors, and time-series and longitudinal data. We demonstrate on several applied problems.

For a d-ary Boolean function Φ: {0, 1}d → {0, 1} and an assignment to its variables x = (x1, x2, . . . , xd) we consider the problem of finding those subsets of the variables that are sufficient to determine the function value with a given probability δ. This is motivated by the task of interpreting predictions of binary classifiers described as Boolean circuits, which can be seen as special cases of neural networks. We show that the problem of deciding whether such subsets of relevant variables of limited size k ≤ d exist is complete for the complexity class NPPP and thus, generally, unfeasible to solve. We then introduce a variant, in which it suffices to check whether a subset determines the function value with probability at least δ or at most δ − γ for 0 < γ < δ. This promise of a probability gap reduces the complexity to the class NPBPP. Finally, we show that finding the minimal set of relevant variables cannot be reasonably approximated, i.e. with an approximation factor d1−α for α > 0, by a polynomial time algorithm unless P = NP. This holds even with the promise of a probability gap.

Nonparametric maximum likelihood estimation is intended to infer the unknown density distribution while making as few assumptions as possible. To alleviate the over parameterization in nonparametric data fitting, smoothing assumptions are usually merged into the estimation. In this paper a novel boosting-based method is introduced to the nonparametric estimation in univariate cases. We deduce the boosting algorithm by the second-order approximation of nonparametric log-likelihood. Gaussian kernel and smooth spline are chosen as weak learners in boosting to satisfy the smoothing assumptions. Simulations and real data experiments demonstrate the efficacy of the proposed approach.

Generative modeling using samples drawn from the probability distribution constitutes a powerful approach for unsupervised machine learning. Quantum mechanical systems can produce probability distributions that exhibit quantum correlations which are difficult to capture using classical models. We show theoretically that such quantum correlations provide a powerful resource for generative modeling. In particular, we provide an unconditional proof of separation in expressive power between a class of widely-used generative models, known as Bayesian networks, and its minimal quantum extension. We show that this expressivity advantage is associated with quantum nonlocality and quantum contextuality. Furthermore, we numerically test this separation on standard machine learning data sets and show that it holds for practical problems. The possibility of quantum advantage demonstrated in this work not only sheds light on the design of useful quantum machine learning protocols but also provides inspiration to draw on ideas from quantum foundations to improve purely classical algorithms.

Some scenarios require the computation of a predictive distribution of a new value evaluated on an objective function conditioned on previous observations. We are interested on using a model that makes valid assumptions on the objective function whose values we are trying to predict. Some of these assumptions may be smoothness or stationarity. Gaussian process (GPs) are probabilistic models that can be interpreted as flexible distributions over functions. They encode the assumptions through covariance functions, making hypotheses about new data through a predictive distribution by being fitted to old observations. We can face the case where several GPs are used to model different objective functions. GPs are non-parametric models whose complexity is cubic on the number of observations. A measure that represents how similar is one GP predictive distribution with respect to another would be useful to stop using one GP when they are modelling functions of the same input space. We are really inferring that two objective functions are correlated, so one GP is enough to model both of them by performing a transformation of the prediction of the other function in case of inverse correlation. We show empirical evidence in a set of synthetic and benchmark experiments that GPs predictive distributions can be compared and that one of them is enough to predict two correlated functions in the same input space. This similarity metric could be extremely useful used to discard objectives in Bayesian many-objective optimization.

Electrochemical impedance spectra is a widely used tool for characterization of fuel cells and electrochemical conversion systems in general. When applied to the on-line monitoring in context of in-field applications, the disturbances, drifts and sensor noise may cause severe distortions in the evaluated spectra, especially in the low-frequency part. Failure to account for the random effects can implicate difficulties in interpreting the spectra and misleading diagnostic reasoning. In the literature, this fact has been largely ignored. In this paper, we propose a computationally efficient approach to the quantification of the spectral uncertainty by quantifying the uncertainty of the equivalent circuit model (ECM) parameters by means of the Variational Bayes (VB) approach. To assess the quality of the VB posterior estimates, we compare the results of VB approach with those obtained with the Markov Chain Monte Carlo (MCMC) algorithm. Namely, MCMC algorithm is expected to return accurate posterior distributions, while VB approach provides the approximative distributions. By using simulated and real data we show that VB approach generates approximations, which although slightly over-optimistic, are still pretty close to the more realistic MCMC estimates. A great advantage of the VB method for online monitoring is low computational load, which is several orders of magnitude lighter than that of MCMC. The performance of VB algorithm is demonstrated on a case of ECM parameters estimation in a 6 cell solid-oxide fuel cell stack. The complete numerical implementation for recreating the results can be found at https://repo.ijs.si/lznidaric/variational-bayes-supplementary-material.

This paper aims to devise a generalized maximum likelihood (ML) estimator to robustly detect signals with unknown noise statistics in multiple-input multiple-output (MIMO) systems. In practice, there is little or even no statistical knowledge on the system noise, which in many cases is non-Gaussian, impulsive and not analyzable. Existing detection methods have mainly focused on specific noise models, which are not robust enough with unknown noise statistics. To tackle this issue, we propose a novel ML detection framework to effectively recover the desired signal. Our framework is a fully probabilistic one that can efficiently approximate the unknown noise distribution through a normalizing flow. Importantly, this framework is driven by an unsupervised learning approach, where only the noise samples are required. To reduce the computational complexity, we further present a low-complexity version of the framework, by utilizing an initial estimation to reduce the search space. Simulation results show that our framework outperforms other existing algorithms in terms of bit error rate (BER) in non-analytical noise environments, while it can reach the ML performance bound in analytical noise environments. The code of this paper is available at https://github.com/skypitcher/manfe.