Conversational recommendation systems can greatly benefit from techniques that explain the reasons behind their actions. We propose techniques that generate explanations by resorting to an auxiliary knowledge graph and an associated knowledge embedding. By exploiting the embedding plausibility score while searching a knowledge graph, we present a method that effectively generates reasons for a recommendation. We then propose a host of techniques to generate balanced reasons both for and against a recommendation, so as to enhance user trust in a conversational recommendation system. To do so, we develop a concrete implementation of Snedegar’s theory of reasons for/against. Experiments at functional, human, and application levels demonstrate that our proposals do improve the interpretability of conversational recommendations systems with controlled computational cost.

Sea-level rise is a well-known consequence of climate change. Several studies have estimated the social and economic impact of the increase in extreme flooding. An efficient way to mitigate its consequences is the development of a flood alert and prediction system, based on high-resolution numerical models and robust sensing networks. However, current models use various simplifying assumptions that compromise accuracy to ensure solvability within a reasonable timeframe, hindering more regular and cost-effective forecasts for various locations along the shoreline. To address these issues, this work proposes a hybrid model for multimodal data processing that combines physics-based numerical simulations, data obtained from a network of sensors, and satellite images to provide refined wave and sea-surface height forecasts, with real results obtained in a critical location within the Port of Santos (the largest port in Latin America). Our approach exhibits faster convergence than data-driven models while achieving more accurate predictions. Moreover, the model handles irregularly sampled time series and missing data without the need for complex preprocessing mechanisms or data imputation while keeping low computational costs through a combination of time encoding, recurrent and graph neural networks. Enabling raw sensor data to be easily combined with existing physics-based models opens up new possibilities for accurate extreme storm tide events forecast systems that enhance community safety and aid policymakers in their decision-making processes.

Question Answering systems based on large language models are widely employed today, benefiting from continuous enhancements and improved performance. The legal domain has become a particularly active focus for Question Answering systems, given its complexity and social importance. This paper offers a discussion on how larger and smaller language models can be used to build a legal document-based Question Answering system. We present a novel model, named Cocoruta, generated by fine-tuning with a corpus of legal documents. In addition, we examine five LLMs as they answer questions related to the legal aspects of a specific domain – the Blue Amazon, a region of particular interest involving environmental issues. The results suggest that while LLMs are not yet of sufficient quality for use as core in legal context Question Answering systems, fine-tuning on specialized corpora imparts a beneficial bias to their legal discourse. Despite having fewer parameters, the Cocoruta model competes well with larger LLMs in this aspect.

We present dPASP, a novel declarative probabilistic logic programming framework that allows for the specification of discrete probabilistic models by neural predicates, relational logic constraints, and interval-valued probabilistic choices. This expressive combination facilitates the construction of models that combine low-level perception (images, texts, etc) and common-sense reasoning, thus providing an excellent tool for neurosymbolic reasoning. To support all such features, we discuss several semantics for probabilistic logic programs that allow one to express nondeterminism, non-monotonic reasoning, contradiction, and (vague) probabilistic knowledge. We also discuss how gradient-based learning can be performed with neural predicates and probabilistic choices under selected semantics. To showcase the possibilities offered by the framework, we present case studies that exploit different semantics and constructs.

Logical Credal Networks or LCNs were recently introduced as a powerful probabilistic logic framework for representing and reasoning with imprecise knowledge. Unlike many existing formalisms, LCNs have the ability to represent cycles and allow specifying marginal and conditional probability bounds on logic formulae which may be important in many realistic scenarios. Previous work on LCNs has focused exclusively on marginal inference, namely computing posterior lower and upper probability bounds on a query formula. In this paper, we explore abductive reasoning tasks such as solving MAP and Marginal MAP queries in LCNs given some evidence. We first formally define the MAP and Marginal MAP tasks for LCNs and subsequently show how to solve these tasks exactly using search-based approaches. We then propose several approximate schemes that allow us to scale MAP and Marginal MAP inference to larger problem instances. An extensive empirical evaluation demonstrates the effectiveness of our algorithms on both random LCN instances as well as LCNs derived from more realistic use-cases.

Credal networks extend Bayesian networks to allow for imprecision in probability values. Marginal MAP is a widely applicable mixed inference task that identifies the most likely assignment for a subset of variables (called MAP variables). However, the task is extremely difficult to solve in credal networks particularly because the evaluation of each complete MAP assignment involves exact likelihood computations (combinatorial sums) over the vertices of a complex joint credal set representing the space of all possible marginal distributions of the MAP variables. In this paper, we explore Credal Marginal MAP inference and develop new exact methods based on variable elimination and depth-first search as well as several approximation schemes based on the mini-bucket partitioning and stochastic local search. An extensive empirical evaluation demonstrates the effectiveness of our new methods on random as well as real-world benchmark problems.

Current research in natural language processing is highly dependent on carefully produced corpora. Most existing resources focus on English; some resources focus on languages such as Chinese and French; few resources deal with more than one language. This paper presents the Pirá dataset, a large set of questions and answers about the ocean and the Brazilian coast both in Portuguese and English. Pirá is, to the best of our knowledge, the first QA dataset with supporting texts in Portuguese, and, perhaps more importantly, the first bilingual QA dataset that includes this language. The Pirá dataset consists of 2261 properly curated question/answer (QA) sets in both languages. The QA sets were manually created based on two corpora: abstracts related to the Brazilian coast and excerpts of United Nation reports about the ocean. The QA sets were validated in a peer-review process with the dataset contributors. We discuss some of the advantages as well as limitations of Pirá, as this new resource can support a set of tasks in NLP such as question-answering, information retrieval, and machine translation.

This demo paper introduces DaMata, a robotjournalist covering deforestation in the Brazilian Amazon.The robot-journalist is based on a pipeline architecture of Natural Language Generation, which yields multilingual daily and monthly reports based on the public data provided by DETER, a real-time deforestation satellite monitor developed and maintained by the Brazilian National Institute for Space Research (INPE).DaMata automatically generates reports in Brazilian Portuguese and English and publishes them on the Twitter platform.Corpus and code are publicly available.1

Probabilistic Answer Set Programming (PASP) combines rules, facts, and independent probabilistic facts. We argue that a very useful modeling paradigm is obtained by adopting a particular semantics for PASP, where one associates a credal set with each consistent program. We examine the basic properties of PASP under this credal semantics, in particular presenting novel results on its complexity and its expressivity, and we introduce an inference algorithm to compute (upper) probabilities given a program.

We analyze the computational complexity of probabilistic logic programming with constraints, disjunctive heads, and aggregates such as sum and max. We consider propositional programs and relational programs with bounded-arity predicates, and look at cautious reasoning (i.e., computing the smallest probability of an atom over all probability models), cautious explanation (i.e., finding an interpretation that maximizes the lower probability of evidence) and cautious maximum-a-posteriori (i.e., finding a partial interpretation for a set of atoms that maximizes their lower probability conditional on evidence) under Lukasiewicz's credal semantics.

Credal networks generalize Bayesian networks to allow for imprecision in probability values. This paper reviews the main results on credal networks under strong independence, as there has been significant progress in the literature during the last decade or so. We focus on computational aspects, summarizing the main algorithms and complexity results for inference and decision making. We address the question "What is really known about strong extensions of credal networks?" by looking at theoretical results and by presenting a short summary of real applications.

We adapt the theory of descriptive complexity to Bayesian networks, to quantify the expressivity of specifications based on predicates and quantifiers. We show that Bayesian network specifications that employ first-order quantification capture the complexity class PP; by allowing quantification over predicates, the resulting Bayesian network specifications capture each class in the hierarchy PP^(NP^...^NP), a result that does not seem to have equivalent in the literature.

Recurrent neural networks are now the state-of-the-art in natural language processing because they can build rich contextual representations and process texts of arbitrary length. However, recent developments on attention mechanisms have equipped feedforward networks with similar capabilities, hence enabling faster computations due to the increase in the number of operations that can be parallelized. We explore this new type of architecture in the domain of question-answering and propose a novel approach that we call Fully Attention Based Information Retriever (FABIR). We show that FABIR achieves competitive results in the Stanford Question Answering Dataset (SQuAD) while having fewer parameters and being faster at both learning and inference than rival methods.

We examine the meaning and the complexity of probabilistic logic programs that consist of a set of rules and a set of independent probabilistic facts (that is, programs based on Sato's distribution semantics). We focus on two semantics, respectively based on stable and on well-founded models. We show that the semantics based on stable models (referred to as the "credal semantics") produces sets of probability measures that dominate infinitely monotone Choquet capacities; we describe several useful consequences of this result. We then examine the complexity of inference with probabilistic logic programs. We distinguish between the complexity of inference when a probabilistic program and a query are given (the inferential complexity), and the complexity of inference when the probabilistic program is fixed and the query is given (the query complexity, akin to data complexity as used in database theory). We obtain results on the inferential and query complexity for acyclic, stratified, and normal propositional and relational programs; complexity reaches various levels of the counting hierarchy and even exponential levels.

Spatial knowledge plays an essential role in human reasoning, permitting tasks such as locating objects in the world (including oneself), reasoning about everyday actions and describing perceptual information. This is also the case in the field of mobile robotics, where one of the most basic (and essential) tasks is the autonomous determination of the pose of a robot with respect to a map, given its perception of the environment. This is the problem of robot self-localisation (or simply the localisation problem). This paper presents a probabilistic algorithm for robot self-localisation that is based on a topological map constructed from the observation of spatial occlusion. Distinct locations on the map are defined by means of a classical formalism for qualitative spatial reasoning, whose base definitions are closer to the human categorisation of space than traditional, numerical, localisation procedures. The approach herein proposed was systematically evaluated through experiments using a mobile robot equipped with a RGB-D sensor. The results obtained show that the localisation algorithm is successful in locating the robot in qualitatively distinct regions.

We examine the inferential complexity of Bayesian networks specified through logical constructs. We first consider simple propositional languages, and then move to relational languages. We examine both the combined complexity of inference (as network size and evidence size are not bounded) and the data complexity of inference (where network size is bounded); we also examine the connection to liftability through domain complexity. Combined and data complexity of several inference problems are presented, ranging from polynomial to exponential classes.

This paper presents initial results towards the development of a logic-based probabilistic event recognition system capable of learning and inferring high-level joint actions from simultaneous task execution demonstrations on a search and rescue scenario. We adopt a probabilistic extension of the Event Calculus defined over Markov Logic Networks (MLN-EC). This formalism was applied to learn and infer the actions of human operators teleoperating robots in a real-world robotic search and rescue task. Experimental results in both simulation and real robots show that the probabilistic event logic can recognise the actions taken by the human teleoperators in real world domains containing two collaborating robots, even with uncertain and noisy data.

Semi-qualitative probabilistic networks (SQPNs) merge two important graphical model formalisms: Bayesian networks and qualitative probabilistic networks. They provide a very general modeling framework by allowing the combination of numeric and qualitative assessments over a discrete domain, and can be compactly encoded by exploiting the same factorization of joint probability distributions that are behind the Bayesian networks. This paper explores the computational complexity of semi-qualitative probabilistic networks, and takes the polytree-shaped networks as its main target. We show that the inference problem is coNP-Complete for binary polytrees with multiple observed nodes. We also show that inferences can be performed in time linear in the number of nodes if there is a single observed node. Because our proof is constructive, we obtain an efficient linear time algorithm for SQPNs under such assumptions. To the best of our knowledge, this is the first exact polynomial-time algorithm for SQPNs. Together these results provide a clear picture of the inferential complexity in polytree-shaped SQPNs.

Automatic classification is one of the basic tasks required in any pattern recognition and human computer interaction application. In this paper, we discuss training probabilistic classifiers with labeled and unlabeled data. We provide a new analysis that shows under what conditions unlabeled data can be used in learning to improve classification performance. We also show that, if the conditions are violated, using unlabeled data can be detrimental to classification performance. We discuss the implications of this analysis to a specific type of probabilistic classifiers, Bayesian networks, and propose a new structure learning algorithm that can utilize unlabeled data to improve classification. Finally, we show how the resulting algorithms are successfully employed in two applications related to human-computer interaction and pattern recognition: facial expression recognition and face detection.

Understanding human emotions is one of the necessary skills for the computer to interact intelligently with human users. The most expressive way humans display emotions is through facial expressions. In this paper, we report on several advances we have made in building a system for classification of facial expressions from continuous video input. We use Bayesian network classifiers for classifying expressions from video. One of the motivating factor in using the Bayesian network classifiers is their ability to handle missing data, both during inference and training. In particular, we are interested in the problem of learning with both labeled and unlabeled data. We show that when using unlabeled data to learn classifiers, using correct modeling assumptions is critical for achieving improved classification performance. Motivated by this, we introduce a classification driven stochastic structure search algorithm for learning the structure of Bayesian network classifiers. We show that with moderate size labeled training sets and large amount of unlabeled data, our method can utilize unlabeled data to improve classification performance. We also provide results using the Naive Bayes (NB) and the Tree-Augmented Naive Bayes (TAN) classifiers, showing that the two can achieve good performance with labeled training sets, but perform poorly when unlabeled data are added to the training set.

We present the formulation and implementation of an image stabilization system capable of stabilizing video with very large displacements between frames. A coarse-to-fine technique is applied in resolution and in model spaces. The registration algorithm uses phase correlation to obtain an initial estimate for translation between images; then Levenberg-Marquardt method for nonlinear optimization is applied to refine the solution. Registration is performed in color space, using a subset of the pixels selected by a gradient-based sub-sampling criterion. This software implementation runs at 5 Hz on non-dedicated hardware (Silicon Graphics R10000 workstation).

Image mosaicing and stabilization can be applied to many areas of industry, such as: surveillance, mapping, teleoperation, maintenance and inspection. The paper gives not only an introduction to key concepts in image mosaicing and stabilization, but also the formulation needed to create real-world systems. We also implemented systems for image mosaicing and stabilization; the implementation and results are also presented.

Light power is affected when it crosses the atmosphere; there is a simple, albeit non-linear, relationship between the radiance of an image at any given wavelength and the distance between object and viewer. This phenomenon is called atmospheric scattering and has been extensively studied by physicists and meteorologists. We present the first analysis of this phenomenon from an image understanding perspective: we investigate a group of techniques for extraction of depth cues solely from the analysis of atmospheric scattering effects in images. Depth from scattering techniques are discussed for indoor and outdoor environments, and experimental tests with real images are presented. We have found that depth cues in outdoor scenes can be recovered with surprising accuracy and can be used as an additional information source for autonomous vehicles.

The paper presents a new application of computer vision to space robotics: a teleoperation interface which analyzes images sent by a mobile robot in space missions and produces position estimates based an the images. The estimates are displayed to the robot operator as additional information to prevent loss of orientation. The current version of the interface detects mountain formations in images and automatically searches for mountain peaks in a given topographic map. A new algorithm for position estimation uses a statistical description of the various disturbances and signals in the measurement process to produce estimates. The authors have tested the system with real images obtained in the Pittsburgh East and Dromedary Peak USGS quadrangles; they report significant improvements in speed and accuracy compared to previous systems.

This paper explores the possibility of using Sun altitude for localization of a robot in totally unknown territory. A set of Sun altitudes is obtained by processing a sequence of time-indexed images of the sky. Each altitude constrains the viewer to a circle on the surface of a celestial body, called the circle of equal altitude. A set of circles of equal altitude can be intersected to yield viewer position. We use this principle to obtain the position on Earth. Since altitude measurements are corrupted by noise, a least-square estimate is numerically calculated from the sequence of altitudes. The paper discusses the necessary theory for Sun-based localization, the technical issues of camera calibration and image processing, and presents preliminary results with real data.

Reliable navigation is critical for a lunar rover, both for autonomous traverses and safeguarded remote teleoperation. This paper describes an implemented system that has autonomously driven a prototype wheeled lunar rover over a kilometer in natural, outdoor terrain. The navigation system uses stereo terrain maps to perform local obstacle avoidance, and arbitrates steering recommendations from both the user and the rover. The paper describes the system architecture, each of the major components, and the experimental results to date.