Artificial Intelligence最新文献

{"title":"“Guess what I'm doing”: Extending legibility to sequential decision tasks","authors":"Miguel Faria ,&nbsp;Francisco S. Melo ,&nbsp;Ana Paiva","doi":"10.1016/j.artint.2024.104107","DOIUrl":"https://doi.org/10.1016/j.artint.2024.104107","url":null,"abstract":"<div><p>In this paper we investigate the notion of <em>legibility</em> in sequential decision tasks under uncertainty. Previous works that extend legibility to scenarios beyond robot motion either focus on deterministic settings or are computationally too expensive. Our proposed approach, dubbed PoLMDP, is able to handle uncertainty while remaining computationally tractable. We establish the advantages of our approach against state-of-the-art approaches in several scenarios of varying complexity. We also showcase the use of our legible policies as demonstrations in machine teaching scenarios, establishing their superiority in teaching new behaviours against the commonly used demonstrations based on the optimal policy. Finally, we assess the legibility of our computed policies through a user study, where people are asked to infer the goal of a mobile robot following a legible policy by observing its actions.</p></div>","PeriodicalId":8434,"journal":{"name":"Artificial Intelligence","volume":"330 ","pages":"Article 104107"},"PeriodicalIF":14.4,"publicationDate":"2024-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140113159","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"aspmc: New frontiers of algebraic answer set counting","authors":"Thomas Eiter ,&nbsp;Markus Hecher ,&nbsp;Rafael Kiesel","doi":"10.1016/j.artint.2024.104109","DOIUrl":"10.1016/j.artint.2024.104109","url":null,"abstract":"<div><p>In the last decade, there has been increasing interest in extensions of answer set programming (ASP) that cater for quantitative information such as weights or probabilities. A wide range of quantitative reasoning tasks for ASP and logic programming, among them probabilistic inference and parameter learning in the neuro-symbolic setting, can be expressed as algebraic answer set counting (AASC) tasks, i.e., weighted model counting for ASP with weights calculated over some semiring, which makes efficient solvers for AASC desirable. In this article, we present <figure><img></figure>, a new solver for AASC that pushes the limits of efficient solvability. Notably, <figure><img></figure> provides improved performance compared to the state of the art in probabilistic inference by exploiting three insights gained from thorough theoretical investigations in our work. Namely, we consider the knowledge compilation step in the AASC pipeline, where the underlying logical theory specified by the answer set program is converted into a tractable circuit representation, on which AASC is feasible in polynomial time. First, we provide a detailed comparison of different approaches to knowledge compilation for programs, revealing that translation to propositional formulas followed by compilation to sd-DNNF seems favorable. Second, we study how the translation to propositional formulas should proceed to result in efficient compilation. This leads to the second and third insight, namely a novel way of breaking the positive cyclic dependencies in a program, called <span><math><msub><mrow><mi>T</mi></mrow><mrow><mi>P</mi></mrow></msub></math></span>-Unfolding, and an improvement to the Clark Completion, the procedure used to transform programs without positive cyclic dependencies into propositional formulas. Both improvements are tailored towards efficient knowledge compilation. Our empirical evaluation reveals that while all three advancements contribute to the success of <figure><img></figure>, <span><math><msub><mrow><mi>T</mi></mrow><mrow><mi>P</mi></mrow></msub></math></span>-Unfolding improves performance significantly by allowing us to handle cyclic instances better.</p></div>","PeriodicalId":8434,"journal":{"name":"Artificial Intelligence","volume":"330 ","pages":"Article 104109"},"PeriodicalIF":14.4,"publicationDate":"2024-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S0004370224000456/pdfft?md5=978459678153434f8adde349d6b98000&pid=1-s2.0-S0004370224000456-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140053605","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Investigating the properties of neural network representations in reinforcement learning","authors":"Han Wang ,&nbsp;Erfan Miahi ,&nbsp;Martha White ,&nbsp;Marlos C. Machado ,&nbsp;Zaheer Abbas ,&nbsp;Raksha Kumaraswamy ,&nbsp;Vincent Liu ,&nbsp;Adam White","doi":"10.1016/j.artint.2024.104100","DOIUrl":"https://doi.org/10.1016/j.artint.2024.104100","url":null,"abstract":"<div><p>In this paper we investigate the properties of representations learned by deep reinforcement learning systems. Much of the early work on representations for reinforcement learning focused on designing fixed-basis architectures to achieve properties thought to be desirable, such as orthogonality and sparsity. In contrast, the idea behind deep reinforcement learning methods is that the agent designer should not encode representational properties, but rather that the data stream should determine the properties of the representation—good representations emerge under appropriate training schemes. In this paper we bring these two perspectives together, empirically investigating the properties of representations that support transfer in reinforcement learning. We introduce and measure six representational properties over more than 25,000 agent-task settings. We consider Deep Q-learning agents with different auxiliary losses in a pixel-based navigation environment, with source and transfer tasks corresponding to different goal locations. We develop a method to better understand <em>why</em> some representations work better for transfer, through a systematic approach varying task similarity and measuring and correlating representation properties with transfer performance. We demonstrate the generality of the methodology by investigating representations learned by a Rainbow agent that successfully transfers across Atari 2600 game modes.</p></div>","PeriodicalId":8434,"journal":{"name":"Artificial Intelligence","volume":"330 ","pages":"Article 104100"},"PeriodicalIF":14.4,"publicationDate":"2024-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S0004370224000365/pdfft?md5=6e885307a80c3c36ff25f169599f1f61&pid=1-s2.0-S0004370224000365-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140051710","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Crossover can guarantee exponential speed-ups in evolutionary multi-objective optimisation","authors":"Duc-Cuong Dang,&nbsp;Andre Opris,&nbsp;Dirk Sudholt","doi":"10.1016/j.artint.2024.104098","DOIUrl":"10.1016/j.artint.2024.104098","url":null,"abstract":"<div><p>Evolutionary algorithms are popular algorithms for multi-objective optimisation (also called Pareto optimisation) as they use a population to store trade-offs between different objectives. Despite their popularity, the theoretical foundation of multi-objective evolutionary optimisation (EMO) is still in its early development. Fundamental questions such as the benefits of the crossover operator are still not fully understood. We provide a theoretical analysis of the well-known EMO algorithms GSEMO and NSGA-II to showcase the possible advantages of crossover: we propose classes of “royal road” functions on which these algorithms cover the whole Pareto front in expected polynomial time if crossover is being used. But when disabling crossover, they require exponential time in expectation to cover the Pareto front. The latter even holds for a large class of black-box algorithms using any elitist selection and any unbiased mutation operator. Moreover, even the expected time to create a single Pareto-optimal search point is exponential. We provide two different function classes, one tailored for one-point crossover and another one tailored for uniform crossover, and we show that some immune-inspired hypermutations cannot avoid exponential optimisation times. Our work shows the first example of an exponential performance gap through the use of crossover for the widely used NSGA-II algorithm and contributes to a deeper understanding of its limitations and capabilities.</p></div>","PeriodicalId":8434,"journal":{"name":"Artificial Intelligence","volume":"330 ","pages":"Article 104098"},"PeriodicalIF":14.4,"publicationDate":"2024-02-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S0004370224000341/pdfft?md5=3dde17723b9eeb078cc6d03de8fc345f&pid=1-s2.0-S0004370224000341-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139994334","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Datalog rewritability and data complexity of ALCHOIQ with closed predicates","authors":"Sanja Lukumbuzya ,&nbsp;Magdalena Ortiz ,&nbsp;Mantas Šimkus","doi":"10.1016/j.artint.2024.104099","DOIUrl":"10.1016/j.artint.2024.104099","url":null,"abstract":"<div><p>We study the relative expressiveness of ontology-mediated queries (OMQs) formulated in the expressive Description Logic <span><math><mi>ALCHOIQ</mi></math></span> extended with closed predicates. In particular, we present a polynomial time translation from OMQs into Datalog with negation under the stable model semantics, the formalism that underlies Answer Set Programming. This is a novel and non-trivial result: the considered OMQs are not only non-monotonic, but also feature a tricky combination of nominals, inverse roles, and counting. We start with atomic queries and then lift our approach to a large class of first-order queries where quantification is “guarded” by closed predicates. Our translation is based on a characterization of the query answering problem via integer programming, and a specially crafted program in Datalog with negation that finds solutions to dynamically generated systems of integer inequalities. As an important by-product of our translation we get that the query answering problem is co-NP-complete in data complexity for the considered class of OMQs. Thus, answering these OMQs in the presence of closed predicates is not harder than answering them in the standard setting. This is not obvious as closed predicates are known to increase data complexity for some existing ontology languages.</p></div>","PeriodicalId":8434,"journal":{"name":"Artificial Intelligence","volume":"330 ","pages":"Article 104099"},"PeriodicalIF":14.4,"publicationDate":"2024-02-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S0004370224000353/pdfft?md5=592b8d7cf49fb907ac6ec3932a493a97&pid=1-s2.0-S0004370224000353-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139943365","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Finding the optimal exploration-exploitation trade-off online through Bayesian risk estimation and minimization","authors":"Stewart Jamieson ,&nbsp;Jonathan P. How ,&nbsp;Yogesh Girdhar","doi":"10.1016/j.artint.2024.104096","DOIUrl":"10.1016/j.artint.2024.104096","url":null,"abstract":"<div><p>We propose <em>endogenous Bayesian risk minimization</em> (EBRM) over policy sets as an approach to online learning across a wide range of settings. Many real-world online learning problems have complexities such as action- and belief-dependent rewards, time-discounting of reward, and heterogeneous costs for actions and feedback; we find that existing online learning heuristics cannot leverage most problem-specific information, to the detriment of their performance. We introduce a belief-space Markov decision process (BMDP) model that can capture these complexities, and further apply the concepts of <em>aleatoric</em>, <em>epistemic</em>, and <em>process</em> risks to online learning. These risk functions describe the risk inherent to the learning problem, the risk due to the agent's lack of knowledge, and the relative quality of its policy, respectively. We demonstrate how computing and minimizing these risk functions guides the online learning agent towards the optimal exploration-exploitation trade-off in any stochastic online learning problem, constituting the basis of the EBRM approach. We also show how Bayes' risk, the minimization objective in stochastic online learning problems, can be decomposed into the aforementioned aleatoric, epistemic, and process risks.</p><p>In simulation experiments, EBRM algorithms achieve state-of-the-art performance across various classical online learning problems, including Gaussian and Bernoulli multi-armed bandits, best-arm identification, mixed objectives with action- and belief-dependent rewards, and dynamic pricing, a finite partial monitoring problem. To our knowledge, it is also the first computationally efficient online learning approach that can provide online bounds on an algorithm's Bayes' risk. Finally, because the EBRM approach is parameterized by a set of policy algorithms, it can be extended to incorporate new developments in online learning algorithms, and is thus well-suited as the foundation for developing real-world learning agents.</p></div>","PeriodicalId":8434,"journal":{"name":"Artificial Intelligence","volume":"330 ","pages":"Article 104096"},"PeriodicalIF":14.4,"publicationDate":"2024-02-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139937811","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Generalized planning as heuristic search: A new planning search-space that leverages pointers over objects","authors":"Javier Segovia-Aguas ,&nbsp;Sergio Jiménez ,&nbsp;Anders Jonsson","doi":"10.1016/j.artint.2024.104097","DOIUrl":"10.1016/j.artint.2024.104097","url":null,"abstract":"<div><p><em>Planning as heuristic search</em> is one of the most successful approaches to <em>classical planning</em> but unfortunately, it does not trivially extend to <em>Generalized Planning</em> (GP); GP aims to compute algorithmic solutions that are valid for a set of classical planning instances from a given domain, even if these instances differ in their number of objects, the initial and goal configuration of these objects and hence, in the number (and possible values) of the state variables. <em>State-space search</em>, as it is implemented by heuristic planners, becomes then impractical for GP. In this paper we adapt the <em>planning as heuristic search</em> paradigm to the generalization requirements of GP, and present the first native heuristic search approach to GP. First, the paper introduces a new pointer-based solution space for GP that is independent of the number of classical planning instances in a GP problem and the size of those instances (i.e. the number of objects, state variables and their domain sizes). Second, the paper defines an upgraded version of our GP algorithm, called <em>Best-First Generalized Planning</em> (<span>BFGP</span>), that implements a <em>best-first search</em> in our pointer-based solution space for GP. Lastly, the paper defines a set of evaluation and heuristic functions for <span>BFGP</span> that assess the structural complexity of the candidate GP solutions, as well as their fitness to a given input set of classical planning instances. The computation of these evaluation and heuristic functions does not require grounding states or actions in advance. Therefore our <em>GP as heuristic search</em> approach can handle large sets of state variables with large numerical domains, e.g. <em>integers</em>.</p></div>","PeriodicalId":8434,"journal":{"name":"Artificial Intelligence","volume":"330 ","pages":"Article 104097"},"PeriodicalIF":14.4,"publicationDate":"2024-02-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.sciencedirect.com/science/article/pii/S000437022400033X/pdfft?md5=155e9598fb32d57f910c200397c5f020&pid=1-s2.0-S000437022400033X-main.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139916222","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Decentralized fused-learner architectures for Bayesian reinforcement learning","authors":"Augustin A. Saucan ,&nbsp;Subhro Das ,&nbsp;Moe Z. Win","doi":"10.1016/j.artint.2024.104094","DOIUrl":"10.1016/j.artint.2024.104094","url":null,"abstract":"<div><p>Decentralized training is a robust solution for learning over an extensive network of distributed agents. Many existing solutions involve the averaging of locally inferred parameters which constrain the architecture to independent agents with identical learning algorithms. Here, we propose decentralized fused-learner architectures for Bayesian reinforcement learning, named fused Bayesian-learner architectures (FBLAs), that are capable of learning an optimal policy by fusing potentially heterogeneous Bayesian policy gradient learners, i.e., agents that employ different learning architectures to estimate the gradient of a control policy. The novelty of FBLAs relies on fusing the full posterior distributions of the local policy gradients. The inclusion of higher-order information, i.e., probabilistic uncertainty, is employed to robustly fuse the locally-trained parameters. FBLAs find the barycenter of all local posterior densities by minimizing the total Kullback–Leibler divergence from the barycenter distribution to the local posterior densities. The proposed FBLAs are demonstrated on a sensor-selection problem for Bernoulli tracking, where multiple sensors observe a dynamic target and only a subset of sensors is allowed to be active at any time.</p></div>","PeriodicalId":8434,"journal":{"name":"Artificial Intelligence","volume":"331 ","pages":"Article 104094"},"PeriodicalIF":14.4,"publicationDate":"2024-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139889913","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Primarily about primaries","authors":"Allan Borodin ,&nbsp;Omer Lev ,&nbsp;Nisarg Shah ,&nbsp;Tyrone Strangway","doi":"10.1016/j.artint.2024.104095","DOIUrl":"https://doi.org/10.1016/j.artint.2024.104095","url":null,"abstract":"<div><p>Much of the social choice literature examines <em>direct</em> voting systems, in which voters submit their ranked preferences over candidates and a voting rule picks a winner. Real-world elections and decision-making processes are often more complex and involve multiple stages. For instance, one popular voting system filters candidates through <em>primaries</em>: first, voters affiliated with each political party vote over candidates of their own party and the voting rule picks a set of candidates, one from each party, who then compete in a general election.</p><p>We present a model to analyze such multi-stage elections, and conduct what is, to the best of our knowledge, the first quantitative comparison of the direct and primary voting systems in terms of the quality of the elected candidate, using the metric of <em>distortion</em>, which attempts to quantify how far from the optimal winner is the actual winner of an election. Our main theoretical result is that voting rules (which are independent of party affiliations, of course) are guaranteed to perform in the primary system within a constant factor of the direct, single stage setting. Surprisingly, the converse does not hold: we show settings in which there exist voting rules that perform significantly better under the primary system than under the direct system. Using simulations, we see that plurality benefits significantly from using a primary system over a direct one, while Condorcet-consistent rules do not.</p></div>","PeriodicalId":8434,"journal":{"name":"Artificial Intelligence","volume":"329 ","pages":"Article 104095"},"PeriodicalIF":14.4,"publicationDate":"2024-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139726522","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Temporal segmentation in multi agent path finding with applications to explainability","authors":"Shaull Almagor ,&nbsp;Justin Kottinger ,&nbsp;Morteza Lahijanian","doi":"10.1016/j.artint.2024.104087","DOIUrl":"10.1016/j.artint.2024.104087","url":null,"abstract":"<div><p>Multi-Agent Path Finding (MAPF) is the problem of planning paths for agents to reach their targets from their start locations, such that the agents do not collide while executing the plan. In many settings, the plan (or a digest thereof) is conveyed to a supervising entity, e.g., for confirmation before execution, for a report, etc. In such cases, we wish to convey that the plan is collision-free with minimal amount of information. To this end, we propose an <em>explanation scheme</em> for MAPF. The scheme decomposes a plan into segments such that within each segment, the agents' paths are disjoint. We can then convey the plan whilst convincing that it is collision-free, using a small number of frames (dubbed an <em>explanation</em>). We can also measure the simplicity of a plan by the number of segments required for the decomposition. We study the complexity of algorithmic problems that arise by the explanation scheme and the tradeoff between the length (makespan) of a plan and its minimal decomposition. We also introduce two centralized (i.e. runs on a single CPU with full knowledge of the multi-agent system) algorithms for planning with explanations. One is based on a coupled search algorithm similar to A^⁎, and the other is a decoupled method based on Conflict-Based Search (CBS). We refer to the latter as <em>Explanation-Guided CBS</em> (XG-CBS), which uses a low-level search for individual agents and maintains a high-level conflict tree to guide the low-level search to avoid collisions as well as increasing the number of segments. We propose four approaches to the low-level search of XG-CBS by modifying A^⁎ for explanations and analyze their effects on the completeness of XG-CBS. Finally, we highlight important aspects of the proposed explanation scheme in various MAPF problems and empirically evaluate the performance of the proposed planning algorithms in a series of benchmark problems.</p></div>","PeriodicalId":8434,"journal":{"name":"Artificial Intelligence","volume":"330 ","pages":"Article 104087"},"PeriodicalIF":14.4,"publicationDate":"2024-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"139832974","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}