Although Deep Neural Networks (DNNs) have great generalization and prediction capabilities, their functioning does not allow a detailed explanation of their behavior. Opaque deep learning models are increasingly used to make important predictions in critical environments, and the danger is that they make and use predictions that cannot be justified or legitimized. Several eXplainable Artificial Intelligence (XAI) methods that separate explanations from machine learning models have emerged, but have shortcomings in faithfulness to the model actual functioning and robustness. As a result, there is a widespread agreement on the importance of endowing Deep Learning models with explanatory capabilities so that they can themselves provide an answer to why a particular prediction was made. First, we address the problem of the lack of universal criteria for XAI by formalizing what an explanation is. We also introduced a set of axioms and definitions to clarify XAI from a mathematical perspective. Finally, we present the Greybox XAI, a framework that composes a DNN and a transparent model thanks to the use of a symbolic Knowledge Base (KB). We extract a KB from the dataset and use it to train a transparent model (i.e., a logistic regression). An encoder-decoder architecture is trained on RGB images to produce an output similar to the KB used by the transparent model. Once the two models are trained independently, they are used compositionally to form an explainable predictive model. We show how this new architecture is accurate and explainable in several datasets.

Communication helps agents to obtain information about others so that better coordinated behavior can be learned. Some existing work communicates predicted future trajectory with others, hoping to get clues about what others would do for better coordination. However, circular dependencies sometimes can occur when agents are treated synchronously so it is hard to coordinate decision-making. In this paper, we propose a novel communication scheme, Sequential Communication (SeqComm). SeqComm treats agents asynchronously (the upper-level agents make decisions before the lower-level ones) and has two communication phases. In negotiation phase, agents determine the priority of decision-making by communicating hidden states of observations and comparing the value of intention, which is obtained by modeling the environment dynamics. In launching phase, the upper-level agents take the lead in making decisions and communicate their actions with the lower-level agents. Theoretically, we prove the policies learned by SeqComm are guaranteed to improve monotonically and converge. Empirically, we show that SeqComm outperforms existing methods in various multi-agent cooperative tasks.

Finite element discretizations of problems in computational physics often rely on adaptive mesh refinement (AMR) to preferentially resolve regions containing important features during simulation. However, these spatial refinement strategies are often heuristic and rely on domain-specific knowledge or trial-and-error. We treat the process of adaptive mesh refinement as a local, sequential decision-making problem under incomplete information, formulating AMR as a partially observable Markov decision process. Using a deep reinforcement learning approach, we train policy networks for AMR strategy directly from numerical simulation. The training process does not require an exact solution or a high-fidelity ground truth to the partial differential equation at hand, nor does it require a pre-computed training dataset. The local nature of our reinforcement learning formulation allows the policy network to be trained inexpensively on much smaller problems than those on which they are deployed. The methodology is not specific to any particular partial differential equation, problem dimension, or numerical discretization, and can flexibly incorporate diverse problem physics. To that end, we apply the approach to a diverse set of partial differential equations, using a variety of high-order discontinuous Galerkin and hybridizable discontinuous Galerkin finite element discretizations. We show that the resultant deep reinforcement learning policies are competitive with common AMR heuristics, generalize well across problem classes, and strike a favorable balance between accuracy and cost such that they often lead to a higher accuracy per problem degree of freedom.

The problem of sampling constrained continuous distributions has frequently appeared in many machine/statistical learning models. Many Monte Carlo Markov Chain (MCMC) sampling methods have been adapted to handle different types of constraints on the random variables. Among these methods, Hamilton Monte Carlo (HMC) and the related approaches have shown significant advantages in terms of computational efficiency compared to other counterparts. In this article, we first review HMC and some extended sampling methods, and then we concretely explain three constrained HMC-based sampling methods, reflection, reformulation, and spherical HMC. For illustration, we apply these methods to solve three well-known constrained sampling problems, truncated multivariate normal distributions, Bayesian regularized regression, and nonparametric density estimation. In this review, we also connect constrained sampling with another similar problem in the statistical design of experiments of constrained design space. Keywords: constrained sampling; Hamilton Monte Carlo; Riemannian Monte Carlo; regularized regression; truncated multivariate Gaussian.

The Inverse Reinforcement Learning (\textit{IRL}) problem has seen rapid evolution in the past few years, with important applications in domains like robotics, cognition, and health. In this work, we explore the inefficacy of current IRL methods in learning an agent's reward function from expert trajectories depicting long-horizon, complex sequential tasks. We hypothesize that imbuing IRL models with structural motifs capturing underlying tasks can enable and enhance their performance. Subsequently, we propose a novel IRL method, SMIRL, that first learns the (approximate) structure of a task as a finite-state-automaton (FSA), then uses the structural motif to solve the IRL problem. We test our model on both discrete grid world and high-dimensional continuous domain environments. We empirically show that our proposed approach successfully learns all four complex tasks, where two foundational IRL baselines fail. Our model also outperforms the baselines in sample efficiency on a simpler toy task. We further show promising test results in a modified continuous domain on tasks with compositional reward functions.

As input distributions evolve over a mission lifetime, maintaining performance of learning-based models becomes challenging. This paper presents a framework to incrementally retrain a model by selecting a subset of test inputs to label, which allows the model to adapt to changing input distributions. Algorithms within this framework are evaluated based on (1) model performance throughout mission lifetime and (2) cumulative costs associated with labeling and model retraining. We provide an open-source benchmark of a satellite pose estimation model trained on images of a satellite in space and deployed in novel scenarios (e.g., different backgrounds or misbehaving pixels), where algorithms are evaluated on their ability to maintain high performance by retraining on a subset of inputs. We also propose a novel algorithm to select a diverse subset of inputs for labeling, by characterizing the information gain from an input using Bayesian uncertainty quantification and choosing a subset that maximizes collective information gain using concepts from batch active learning. We show that our algorithm outperforms others on the benchmark, e.g., achieves comparable performance to an algorithm that labels 100% of inputs, while only labeling 50% of inputs, resulting in low costs and high performance over the mission lifetime.

Bandits with knapsacks (BwK) is an influential model of sequential decision-making under uncertainty that incorporates resource consumption constraints. In each round, the decision-maker observes an outcome consisting of a reward and a vector of nonnegative resource consumptions, and the budget of each resource is decremented by its consumption. In this paper we introduce a natural generalization of the stochastic BwK problem that allows non-monotonic resource utilization. In each round, the decision-maker observes an outcome consisting of a reward and a vector of resource drifts that can be positive, negative or zero, and the budget of each resource is incremented by its drift. Our main result is a Markov decision process (MDP) policy that has constant regret against a linear programming (LP) relaxation when the decision-maker knows the true outcome distributions. We build upon this to develop a learning algorithm that has logarithmic regret against the same LP relaxation when the decision-maker does not know the true outcome distributions. We also present a reduction from BwK to our model that shows our regret bound matches existing results.

Our implementation of recurrent model-free RL outperforms the on-policy version (PPO/A2C-GRU), and a recent model-based POMDP algorithm (VRM) on most tasks of a POMDP benchmark where VRM was evaluated in their paper. While algorithms for decision-making typically focus on relatively easy problems where everything is known, most realistic problems involve noise and incomplete information. Complex algorithms have been proposed to tackle these complex problems, but there's a simple approach that (in theory) works on both the easy and the complex problems. We show how to make this simple approach work in practice. Decision-making tasks in the real world are messy, with noise, occlusions, and uncertainty that are typically missing from their canonical problem formulation as a Markov decision process (MDP; Bellman, 1957).

We introduce a simulation-based, amortised Bayesian inference scheme to infer the parameters of random walks. Our approach learns the posterior distribution of the walks' parameters with a likelihood-free method. In the first step a graph neural network is trained on simulated data to learn optimized low-dimensional summary statistics of the random walk. In the second step an invertible neural network generates the posterior distribution of the parameters from the learnt summary statistics using variational inference. We apply our method to infer the parameters of the fractional Brownian motion model from single trajectories. The computational complexity of the amortized inference procedure scales linearly with trajectory length, and its precision scales similarly to the Cram{\'e}r-Rao bound over a wide range of lengths. The approach is robust to positional noise, and generalizes well to trajectories longer than those seen during training. Finally, we adapt this scheme to show that a finite decorrelation time in the environment can furthermore be inferred from individual trajectories.

Forecast combinations have flourished remarkably in the forecasting community and, in recent years, have become part of the mainstream of forecasting research and activities. Combining multiple forecasts produced from single (target) series is now widely used to improve accuracy through the integration of information gleaned from different sources, thereby mitigating the risk of identifying a single "best" forecast. Combination schemes have evolved from simple combination methods without estimation, to sophisticated methods involving time-varying weights, nonlinear combinations, correlations among components, and cross-learning. They include combining point forecasts and combining probabilistic forecasts. This paper provides an up-to-date review of the extensive literature on forecast combinations, together with reference to available open-source software implementations. We discuss the potential and limitations of various methods and highlight how these ideas have developed over time. Some important issues concerning the utility of forecast combinations are also surveyed. Finally, we conclude with current research gaps and potential insights for future research.