Solar irradiance forecasts can be dynamic and unreliable due to changing weather conditions. Near the Arctic circle, this also translates into a distinct set of further challenges. This work is forecasting solar irradiance with Norwegian data using variations of Long-Short-Term Memory units (LSTMs). In order to gain more trustworthiness of results, the probabilistic approaches Quantile Regression (QR) and Maximum Likelihood (MLE) are optimized on top of the LSTMs, providing measures of uncertainty for the results. MLE is further extended by using a Johnson's SU distribution, a Johnson's SB distribution, and a Weibull distribution in addition to a normal Gaussian to model parameters. Contrary to a Gaussian, Weibull, Johnson's SU and Johnson's SB can return skewed distributions, enabling it to fit the non-normal solar irradiance distribution more optimally. The LSTMs are compared against each other, a simple Multi-layer Perceptron (MLP), and a smart-persistence estimator. The proposed LSTMs are found to be more accurate than smart persistence and the MLP for a multi-horizon, day-ahead (36 hours) forecast. The deterministic LSTM showed better root mean squared error (RMSE), but worse mean absolute error (MAE) than a MLE with Johnson's SB distribution. Probabilistic uncertainty estimation is shown to fit relatively well across the distribution of observed irradiance. While QR shows better uncertainty estimation calibration, MLE with Johnson's SB, Johnson's SU, or Gaussian show better performance in the other metrics employed. Optimizing and comparing the models against each other reveals a seemingly inherent trade-off between point-prediction and uncertainty estimation calibration.

Symmetries have proven useful in machine learning models, improving generalisation and overall performance. At the same time, recent advancements in learning dynamical systems rely on modelling the underlying Hamiltonian to guarantee the conservation of energy. These approaches can be connected via a seminal result in mathematical physics: Noether's theorem, which states that symmetries in a dynamical system correspond to conserved quantities. This work uses Noether's theorem to parameterise symmetries as learnable conserved quantities. We then allow conserved quantities and associated symmetries to be learned directly from train data through approximate Bayesian model selection, jointly with the regular training procedure. As training objective, we derive a variational lower bound to the marginal likelihood. The objective automatically embodies an Occam's Razor effect that avoids collapse of conservation laws to the trivial constant, without the need to manually add and tune additional regularisers. We demonstrate a proof-ofprinciple on n-harmonic oscillators and n-body systems. We find that our method correctly identifies the correct conserved quantities and U(n) and SE(n) symmetry groups, improving overall performance and predictive accuracy on test data.

Simulation-based inference (SBI) is the preferred framework for estimating parameters of intractable models in science and engineering. A significant challenge in this context is the large computational cost of simulating data from complex models, and the fact that this cost often depends on parameter values. We therefore propose \textit{cost-aware SBI methods} which can significantly reduce the cost of existing sampling-based SBI methods, such as neural SBI and approximate Bayesian computation. This is achieved through a combination of rejection and self-normalised importance sampling, which significantly reduces the number of expensive simulations needed. Our approach is studied extensively on models from epidemiology to telecommunications engineering, where we obtain significant reductions in the overall cost of inference.

We study the sample complexity of the plug-in approach for learning $\varepsilon$-optimal policies in average-reward Markov decision processes (MDPs) with a generative model. The plug-in approach constructs a model estimate then computes an average-reward optimal policy in the estimated model. Despite representing arguably the simplest algorithm for this problem, the plug-in approach has never been theoretically analyzed. Unlike the more well-studied discounted MDP reduction method, the plug-in approach requires no prior problem information or parameter tuning. Our results fill this gap and address the limitations of prior approaches, as we show that the plug-in approach is optimal in several well-studied settings without using prior knowledge. Specifically it achieves the optimal diameter- and mixing-based sample complexities of $\widetilde{O}\left(SA \frac{D}{\varepsilon^2}\right)$ and $\widetilde{O}\left(SA \frac{\tau_{\mathrm{unif}}}{\varepsilon^2}\right)$, respectively, without knowledge of the diameter $D$ or uniform mixing time $\tau_{\mathrm{unif}}$. We also obtain span-based bounds for the plug-in approach, and complement them with algorithm-specific lower bounds suggesting that they are unimprovable. Our results require novel techniques for analyzing long-horizon problems which may be broadly useful and which also improve results for the discounted plug-in approach, removing effective-horizon-related sample size restrictions and obtaining the first optimal complexity bounds for the full range of sample sizes without reward perturbation.

Virtually any model we use in machine learning to make predictions does not perfectly represent reality. So, most of the learning happens under model misspecification. In this work, we present a novel analysis of the generalization performance of Bayesian model averaging under model misspecification and i.i.d. This analysis shows, in simple and intuitive terms, that Bayesian model averaging provides suboptimal generalization performance when the model is misspecified. In consequence, we provide strong theoretical arguments showing that Bayesian methods are not optimal for learning predictive models, unless the model class is perfectly specified.

The log-likelihood of a generative model often involves both positive and negative terms. As a result, maximum likelihood estimation is expensive. We show how to instead apply a version of noise-contrastive estimation---a general parameter estimation method with a less expensive stochastic objective. Our specific instantiation of this general idea works out in an interestingly non-trivial way and has provable guarantees for its optimality, consistency and efficiency. On several synthetic and real-world datasets, our method shows benefits: for the model to achieve the same level of log-likelihood on held-out data, our method needs considerably fewer function evaluations and less wall-clock time.

Conventional imitation learning assumes access to the actions of demonstrators, but these motor signals are often non-observable in naturalistic settings. To address these challenges, we explore deep generative modeling of state-only sequences with non-Markov Decision Process (nMDP), where the policy is an energy-based prior in the latent space of the state transition generator. We develop maximum likelihood estimation to achieve model-based imitation, which involves short-run MCMC sampling from the prior and importance sampling for the posterior. The learned model enables \textit{decision-making as inference}: model-free policy execution is equivalent to prior sampling, model-based planning is posterior sampling initialized from the policy. We demonstrate the efficacy of the proposed method in a prototypical path planning task with non-Markovian constraints and show that the learned model exhibits strong performances in challenging domains from the MuJoCo suite.

Modern CNN-based object detectors assign anchors for ground-truth objects under the restriction of object-anchor Intersection-over-Unit (IoU). In this study, we propose a learning-to-match approach to break IoU restriction, allowing objects to match anchors in a flexible manner. Our approach, referred to as FreeAnchor, updates hand-crafted anchor assignment to "free" anchor matching by formulating detector training as a maximum likelihood estimation (MLE) procedure. FreeAnchor targets at learning features which best explain a class of objects in terms of both classification and localization. FreeAnchor is implemented by optimizing detection customized likelihood and can be fused with CNN-based detectors in a plug-and-play manner.

The key distinguishing property of a Bayesian approach is marginalization, rather than using a single setting of weights. Bayesian marginalization can particularly improve the accuracy and calibration of modern deep neural networks, which are typically underspecified by the data, and can represent many compelling but different solutions. We show that deep ensembles provide an effective mechanism for approximate Bayesian marginalization, and propose a related approach that further improves the predictive distribution by marginalizing within basins of attraction, without significant overhead. We also investigate the prior over functions implied by a vague distribution over neural network weights, explaining the generalization properties of such models from a probabilistic perspective. From this perspective, we explain results that have been presented as mysterious and distinct to neural network generalization, such as the ability to fit images with random labels, and show that these results can be reproduced with Gaussian processes.

Training Restricted Boltzmann Machines (RBMs) has been challenging for a long time due to the difficulty of computing precisely the log-likelihood gradient. Over the past decades, many works have proposed more or less successful recipes but without studying systematically the crucial quantity of the problem: the mixing time i.e. the number of MCMC iterations needed to sample completely new configurations from a model. In this work, we show that this mixing time plays a crucial role in the behavior and stability of the trained model, and that RBMs operate in two well-defined distinct regimes, namely equilibrium and out-of-equilibrium, depending on the interplay between this mixing time of the model and the number of MCMC steps, k, used to approximate the gradient. We further show empirically that this mixing time increases along the learning, which often implies a transition from one regime to another as soon as k becomes smaller than this time.In particular, we show that using the popular k (persistent) contrastive divergence approaches, with k small, the dynamics of the fitted model are extremely slow and often dominated by strong out-of-equilibrium effects. On the contrary, RBMs trained in equilibrium display much faster dynamics, and a smooth convergence to dataset-like configurations during the sampling.Finally, we discuss how to exploit in practice both regimes depending on the task one aims to fulfill: (i) short k s can be used to generate convincing samples in short learning times, (ii) large k (or increasingly large) must be used to learn the correct equilibrium distribution of the RBM.