Showing papers in "Artificial Intelligence in 2022"

TL;DR: In this article, the authors provide an extensive overview of the use of knowledge graphs in the context of explainable machine learning, and provide an analytical framework to explore the current landscape of Explainable Machine Learning.

TL;DR: This paper investigates gradual semantics dealing with weighted graphs, a family of graphs where each argument has an initial weight and may be attacked by other arguments, and proposes three novel semantics that are very efficient in that they compute the strengths of arguments in less than 20 iterations and in a very short time.

TL;DR: The chapters in the book illustrate how machine learning and deep learning algorithms and models are designed, optimized, and deployed.

TL;DR: Novel Bayesian variable selection methods, targeting feature interaction selection for factorization machines, which effectively reduce the number of interactions are proposed.

TL;DR: A mechanism for parsimonious eXplainable AI (XAI) is proposed and HAExA, a human-agent architecture allowing to make it operational for remote robots is proposed, which relies on both contrastive explanations and explanation filtering.

TL;DR: The history of the Winograd Schema Challenge is reviewed, and a number of AI systems, based on large pre-trained transformer-based language models and fine-tuned on these kinds of problems, achieved better than 90% accuracy.

TL;DR: The first Continual Learning in Computer Vision Challenge (CLCVC) as discussed by the authors was held in 2019, which was one of the first opportunities to evaluate different continual learning algorithms on a common hardware with a large set of shared evaluation metrics and 3 different settings based on the realistic CORe50 video benchmark.

TL;DR: The m A ⁎ action language as mentioned in this paper is a generalization of the single-agent action languages to the case of multi-agent domains, which allows the representation of different types of actions that an agent can perform in a domain where many other agents might be present.

TL;DR: Wang et al. as mentioned in this paper proposed a neural network-based end-to-end event coreference architecture (E 3 C ) that jointly models event detection and resolution tasks and learn to extract features from raw text automatically.

TL;DR: From this, the first formal causal discovery algorithm for discovering agents from empirical data is derived, and algorithms for translating between causal models and game-theoretic influence diagrams are given.

TL;DR: This work proposes a framework for dimensionality estimation and reconstruction of multiple noisy manifolds embedded in a noisy environment and demonstrates the workings of the framework on two synthetic data sets, presenting challenging features for state-of-the-art techniques in Multi-Manifold learning.

TL;DR: This work shows that it can leverage the wide literature on the Situation Calculus and ConGolog programs to formalise this kind of manufacturing, and investigates how to synthesize process plan controllers in this first-order state setting.

TL;DR: In this paper, a POMC Pareto optimization approach is proposed to solve the submodular optimization problem for function f with constraint bound B that changes over time, where α f is the sub-modularity ratio of f and B is the constraint bound.

TL;DR: In this article, the authors consider a model where the decision maker receives a noisy estimate of each candidate's quality, whose variance depends on the candidate's group, and show that such differential variance is a key feature of many selection problems.

TL;DR: Wang et al. as discussed by the authors proposed a novel Framework of Integrating both Latent and Explicit features (FILE), to better deal with the no-relation status and hence improve the overall trust/distrust prediction performance.

TL;DR: In this paper, the authors introduce the notion of margin of victory (MoV) for tournament solutions, which is a robustness measure for individual alternatives: for winners, the MoV captures the distance from dropping out of the winner set, and for non-winners, the distance to entering the set, measured in terms of which pairwise comparisons would have to be reversed in order to achieve the desired outcome.

TL;DR: In this paper, it was shown that the many-valued extension of first-order logic over databases with incomplete information represented by null values is no more powerful than the usual two-valued logic with the standard Boolean interpretation of the connectives.

TL;DR: In this article, the authors consider the problem of fair allocation of indivisible goods when maximin share is used as the measure of fairness and give constant approximation guarantees for agents with submodular and XOS valuations, and a logarithmic bound for the case of subadditive valuations.

TL;DR: In this paper, the authors analyzed the impact of the cutoff time κ (the time spent evaluating a configuration for a problem instance) on the expected number of configuration comparisons required to find the optimal parameter value for the performance metrics (the measure used to judge the performance of a configuration).

TL;DR: In this article, the authors address the task of synthesizing plans that necessitate reasoning about the beliefs of other agents, using a single agent with the potential for goals and actions that involve nested beliefs, non-homogeneous agents, and the ability for one agent to reason as if it were another.

TL;DR: In this paper, the authors study the problem of designing an auction design for a seller to sell a single commodity in a social network, where each individual (the seller or a buyer) can only communicate with her neighbors.

TL;DR: In this article , a case study applying learning-based distributionally robust model predictive control to highway motion planning under stochastic uncertainty of the lane change behavior of surrounding road users is presented.

TL;DR: In this paper, the authors proposed an end point search (EPS) algorithm for the Euclidean Shortest Path Problem (ESPP) in two dimensions, which is search-free, simultaneously fast, and returns a path within a fixed bound of the optimal solution.

TL;DR: This manuscript proposes two Bayesian probabilistic generative models of networks, whose novelty consists in the interrelationship of overlapping communities, roles, their behavioral patterns and node attributes.

TL;DR: A method to detect implicit model patterns (such as global constraints) that might be able to replace parts of a combinatorial problem model that are expressed at a low-level that can help non-expert users write higher-level models that are easier to reason about and often yield better performance.

TL;DR: In this paper, the authors propose a compact encoding for representing and reasoning about the outcomes of non-transferable utility coalitional games based on answer set programming, which can succinctly encode several games of interest within a wide range of application domains.

TL;DR: In this article , an algorithm for inverse reinforcement learning (IRL) in partially observable Markov decision processes (POMDPs) is proposed to solve the intrinsic nonconvexity of the forward problem in a scalable manner through a sequential linear programming scheme.

TL;DR: In this article, a large collection of logical operators from the fuzzy logic literature are studied in a differentiable learning setting, and it is shown that many of these operators, including some of the most well known, are highly unsuitable in this setting.

TL;DR: A coherent and flexible framework for modelling abstraction (and abstraction-like) methods based on graph transformations is presented and used to identify and investigate connections between refinement and heuristics—two concepts that have usually been considered as unrelated in the literature.

TL;DR: The research on computer vision systems has been increasing every day and has led to the design of multiple types of these systems with innumerous applications in our daily life as discussed by the authors , and the recent advances in artificial intelligence, together with the huge amount of digital visual data now available, have boosted vision system performance in several ways.