Leveraging well-established MCMC strategies, we propose MCMC-interactive variational inference (MIVI) to not only estimate the posterior in a time constrained manner, but also facilitate the design of MCMC transitions. Constructing a variational distribution followed by a short Markov chain that has parameters to learn, MIVI takes advantage of the complementary properties of variational inference and MCMC to encourage mutual improvement. On one hand, with the variational distribution locating high posterior density regions, the Markov chain is optimized within the variational inference framework to efficiently target the posterior despite a small number of transitions. On the other hand, the optimized Markov chain with considerable flexibility guides the variational distribution towards the posterior and alleviates its underestimation of uncertainty. Furthermore, we prove the optimized Markov chain in MIVI admits extrapolation, which means its marginal distribution gets closer to the true posterior as the chain grows. Therefore, the Markov chain can be used separately as an efficient MCMC scheme. Experiments show that MIVI not only accurately and efficiently approximates the posteriors but also facilitates designs of stochastic gradient MCMC and Gibbs sampling transitions.

This work demonstrates how a complete statistical model quantifying the evolution of pairs of aligned proteins can be constructed from a time-parameterised substitution matrix and a time-parameterised 3-state alignment machine. All parameters of such a model can be inferred from any benchmark data-set of aligned protein sequences. This allows us to examine nine well-known substitution matrices on six benchmarks curated using various structural alignment methods; any matrix that does not explicitly model a "time"-dependent Markov process is converted to a corresponding base-matrix that does. In addition, a new optimal matrix is inferred for each of the six benchmarks. Using Minimum Message Length (MML) inference, all 15 matrices are compared in terms of measuring the Shannon information content of each benchmark. This has resulted in a new and clear overall best performed time-dependent Markov matrix, MMLSUM, and its associated 3-state machine, whose properties we have analysed in this work. For standard use, the MMLSUM series of (log-odds) \textit{scoring} matrices derived from the above Markov matrix, are available at https://lcb.infotech.monash.edu.au/mmlsum.

Many common loss functions such as mean-squared-error, cross-entropy, and reconstruction loss are unnecessarily rigid. Under a probabilistic interpretation, these common losses correspond to distributions with fixed shapes and scales. We instead argue for optimizing full likelihoods that include parameters like the normal variance and softmax temperature. Joint optimization of these "likelihood parameters" with model parameters can adaptively tune the scales and shapes of losses in addition to the strength of regularization. We explore and systematically evaluate how to parameterize and apply likelihood parameters for robust modeling, outlier-detection, and re-calibration. Additionally, we propose adaptively tuning $L_2$ and $L_1$ weights by fitting the scale parameters of normal and Laplace priors and introduce more flexible element-wise regularizers.

Meta-learning is a powerful tool for learning policies that can adapt efficiently when deployed in new tasks. If however the meta-training tasks have sparse rewards, the need for exploration during meta-training is exacerbated given that the agent has to explore and learn across many tasks. We show that current meta-learning methods can fail catastrophically in such environments. To address this problem, we propose HyperX, a novel method for meta-learning in sparse reward tasks. Using novel reward bonuses for meta-training, we incentivise the agent to explore in approximate hyper-state space, i.e., the joint state and approximate belief space, where the beliefs are over tasks. We show empirically that these bonuses allow an agent to successfully learn to solve sparse reward tasks where existing meta-learning methods fail.

ParaMonte::Python (standing for Parallel Monte Carlo in Python) is a serial and MPI-parallelized library of (Markov Chain) Monte Carlo (MCMC) routines for sampling mathematical objective functions, in particular, the posterior distributions of parameters in Bayesian modeling and analysis in data science, Machine Learning, and scientific inference in general. In addition to providing access to fast high-performance serial/parallel Monte Carlo and MCMC sampling routines, the ParaMonte::Python library provides extensive post-processing and visualization tools that aim to automate and streamline the process of model calibration and uncertainty quantification in Bayesian data analysis. Furthermore, the automatically-enabled restart functionality of ParaMonte::Python samplers ensure seamless fully-deterministic into-the-future restart of Monte Carlo simulations, should any interruptions happen. The ParaMonte::Python library is MIT-licensed and is permanently maintained on GitHub at https://github.com/cdslaborg/paramonte/tree/master/src/interface/Python.

Background: Catastrophic forgetting is the notorious vulnerability of neural networks to the changes in the data distribution during learning. This phenomenon has long been considered a major obstacle for using learning agents in realistic continual learning settings. A large body of continual learning research assumes that task boundaries are known during training. However, only a few works consider scenarios in which task boundaries are unknown or not well defined -- task agnostic scenarios. The optimal Bayesian solution for this requires an intractable online Bayes update to the weights posterior. Contributions: We aim to approximate the online Bayes update as accurately as possible. To do so, we derive novel fixed-point equations for the online variational Bayes optimization problem, for multivariate Gaussian parametric distributions. By iterating the posterior through these fixed-point equations, we obtain an algorithm (FOO-VB) for continual learning which can handle non-stationary data distribution using a fixed architecture and without using external memory (i.e. without access to previous data). We demonstrate that our method (FOO-VB) outperforms existing methods in task agnostic scenarios. FOO-VB Pytorch implementation will be available online.

We propose to distinguish between deterministic conditioning, that is, conditioning on a sample from the joint data distribution, and stochastic conditioning, that is, conditioning on the distribution of the observable variable. Mostly, probabilistic programs follow the Bayesian approach by choosing a prior distribution of parameters and conditioning on observations. In a basic setting, individual observations are In a basic setting, individual observations are samples from the joint data distribution. However, observations may also be independent samples from marginal data distributions of each observable variable, summary statistics, or even data distributions themselves . These cases naturally appear in real life scenarios: samples from marginal distributions arise when different observations are collected by different parties, summary statistics (mean, variance, and quantiles) are often used to represent data collected over a large population, and data distributions may represent uncertainty during inference about future states of the world, that is, in planning. Probabilistic programming languages and frameworks which support conditioning on samples from the joint data distribution are not directly capable of expressing such models. We define the notion of stochastic conditioning and describe extensions of known general inference algorithms to probabilistic programs with stochastic conditioning. In case studies we provide probabilistic programs for several problems of statistical inference which are impossible or difficult to approach otherwise, perform inference on the programs, and analyse the results.

This book is devoted to the problem of sequential probability forecasting, that is, predicting the probabilities of the next outcome of a growing sequence of observations given the past. This problem is considered in a very general setting that unifies commonly used probabilistic and non-probabilistic settings, trying to make as few as possible assumptions on the mechanism generating the observations. A common form that arises in various formulations of this problem is that of mixture predictors, which are formed as a combination of a finite or infinite set of other predictors attempting to combine their predictive powers. The main subject of this book are such mixture predictors, and the main results demonstrate the universality of this method in a very general probabilistic setting, but also show some of its limitations. While the problems considered are motivated by practical applications, involving, for example, financial, biological or behavioural data, this motivation is left implicit and all the results exposed are theoretical. The book targets graduate students and researchers interested in the problem of sequential prediction, and, more generally, in theoretical analysis of problems in machine learning and non-parametric statistics, as well as mathematical and philosophical foundations of these fields. The material in this volume is presented in a way that presumes familiarity with basic concepts of probability and statistics, up to and including probability distributions over spaces of infinite sequences. Familiarity with the literature on learning or stochastic processes is not required.

We give methods for Bayesian inference of directed acyclic graphs, DAGs, and the induced causal effects from passively observed complete data. Our methods build on a recent Markov chain Monte Carlo scheme for learning Bayesian networks, which enables efficient approximate sampling from the graph posterior, provided that each node is assigned a small number K of candidate parents. We present algorithmic tricks to significantly reduce the space and time requirements of the method, making it feasible to use substantially larger values of K. Furthermore, we investigate the problem of selecting the candidate parents per node so as to maximize the covered posterior mass. Finally, we combine our sampling method with a novel Bayesian approach for estimating causal effects in linear Gaussian DAG models. Numerical experiments demonstrate the performance of our methods in detecting ancestor-descendant relations, and in effect estimation our Bayesian method is shown to outperform existing approaches.

Reinforcement learning methods for traffic signal control has gained increasing interests recently and achieved better performances compared with traditional transportation methods. However, reinforcement learning based methods usually requires heavy training data and computational resources which largely limit its application in real-world traffic signal control. This makes meta-learning, which enables data-efficient and fast-adaptation training by leveraging the knowledge of previous learning experiences, catches attentions in traffic signal control. In this paper, we propose a novel value-based Bayesian meta-reinforcement learning framework BM-DQN to robustly speed up the learning process in new scenarios by utilizing well-trained prior knowledge learned from existing scenarios. This framework based on our proposed fast-adaptation variation to Gradient-EM Bayesian Meta-learning and the fast update advantage of DQN, which allows fast adaptation to new scenarios with continual learning ability and robustness to uncertainty. The experiments on 2D navigation and traffic signal control show that our proposed framework adapts more quickly and robustly in new scenarios than previous methods, and specifically, much better continual learning ability in heterogeneous scenarios.