We develop a computational approach to Metric Answer Set programming (ASP) to allow for expressing quantitative temporal constrains, like durations and deadlines. A central challenge is to maintain scalability when de...
ISBN:
(纸本)9783031742088;9783031742095
We develop a computational approach to Metric Answer Set programming (ASP) to allow for expressing quantitative temporal constrains, like durations and deadlines. A central challenge is to maintain scalability when dealing with fine-grained timing constraints, which can significantly exacerbate ASP's grounding bottleneck. To address this issue, we leverage extensions of ASP with difference constraints, a simplified form of linear constraints, to handle time-related aspects externally. Our approach effectively decouples metric ASP from the granularity of time, resulting in a solution that is unaffected by time precision.
Weather forecasting is important for saving lives, protecting property, and supporting economic activities. It provides timely warnings for severe weather, improves agricultural planning, and aids in disaster manageme...
详细信息
ISBN:
(纸本)9783031742088;9783031742095
Weather forecasting is important for saving lives, protecting property, and supporting economic activities. It provides timely warnings for severe weather, improves agricultural planning, and aids in disaster management. Neural networks and deep learning methods can achieve impressive accuracy in weather prediction, but their black-box nature lacks in explainability. To address this limitation, we investigated the potential of FastLAS, an Inductive logicprogramming (ILP) framework, to produce reliable and, more important, explainable weather predictions. FastLAS learns ASP programs whose syntax and structural semantics resemble natural human language, making them easily understandable and interpretable by humans. the supportedness of stable models allows a clear explanation of the predictions. Our empirical evaluation on data from an Italian weather forecasting center shows that our approach is capable of learning predictive models from small dataset (a few samples instead of the thousands needed by neural networks) achieving an accuracy higher than statistical machine learning base lines.
this volume contains the refereed proceedings of the 12thinternationalconference on logicprogramming and nonmonotonicreasoning, LPNMR 2013, held in September 2013 in Corunna, Spain. the 34 revised full papers (22 ...
ISBN:
(数字)9783642405648
ISBN:
(纸本)9783642405631;9783642405648
this volume contains the refereed proceedings of the 12thinternationalconference on logicprogramming and nonmonotonicreasoning, LPNMR 2013, held in September 2013 in Corunna, Spain. the 34 revised full papers (22 technical papers, 9 application description, and 3 system descriptions) and 19 short papers (11 technical papers, 3 application descriptions, and 5 system descriptions) presented together with 2 invited talks, were carefully reviewed and selected from 91 submissions. Being a forum for exchanging ideas on declarative logicprogramming, nonmonotonicreasoning, and knowledge representation, the conference aims to facilitate interactions between those researchers and practitioners interested in the design and implementation of logic-based programming languages and database systems, and those who work in the area of knowledge representation and nonmonotonicreasoning.
Large Language Models (LLMs) lack the ability for commonsense reasoning and learning from text. In this work, we present a system, called LLM2LAS, for learning commonsense knowledge from story-based question and answe...
详细信息
ISBN:
(纸本)9783031742088;9783031742095
Large Language Models (LLMs) lack the ability for commonsense reasoning and learning from text. In this work, we present a system, called LLM2LAS, for learning commonsense knowledge from story-based question and answering expressed in natural language. LLM2LAS combines the semantic parsing capability of LLMs with ILASP for learning commonsense knowledge expressed as answer set programs. LLM2LAS requires only few examples of questions and answers to learn general commonsense knowledge and correctly answer unseen questions. An empirical evaluation demonstrates the viability of our approach.
We develop an approach to solve curriculum-based course timetabling (CB-CTT) problems with Large Neighborhood Prioritized Search (LNPS) based on Answer Set programming (ASP). LNPS is a metaheuristic that starts with a...
详细信息
ISBN:
(纸本)9783031742088;9783031742095
We develop an approach to solve curriculum-based course timetabling (CB-CTT) problems with Large Neighborhood Prioritized Search (LNPS) based on Answer Set programming (ASP). LNPS is a metaheuristic that starts with an initial solution and then iteratively tries to obtain improved solutions by alternately destroying and prioritized searching for a current solution. Our approach relies on high-level domain-specific LNPS configurations for efficient CB-CTT solving, and boththe grounding and solving tasks are delegated to a heuristically-driven answer set optimizer implementing the LNPS algorithm. the resulting system teaspoon-lnps demonstrates that LNPS can significantly enhance the solving performance of ASP for CB-CTT solving. Furthermore, we show the competitiveness of our declarative approach by empirically contrasting it to the best known bounds obtained by more dedicated algorithms.
the analysis of properties of consequence operators has been a very active field in the formative years of non-monotonic reasoning. One possible approach to do this is to start with a model-theoretic semantics and the...
详细信息
ISBN:
(纸本)9783031742088;9783031742095
the analysis of properties of consequence operators has been a very active field in the formative years of non-monotonic reasoning. One possible approach to do this is to start with a model-theoretic semantics and then to study the logical consequence relation induced by that semantics. In this paper we follow that approach and analyse resulting consequence operators of so-called characterization logics. Roughly speaking, a characterization logic characterizes, via its own notion of ordinary equivalence, another logic's notion of strong equivalence. For example, the logic of here and there is a characterization logic for answer set programs, because strong equivalence of the latter is characterized by ordinary equivalence of the former. In previous work, we showed that the consideration of finite knowledge bases only - a common assumption in the field of knowledge representation - guarantees the existence (and uniqueness) of characterization logics. In this paper, we apply this existence result to the field of abstract argumentation. We show that the associated consequence operator outputs a so-called reverse kernel, a useful construct that received comparably little attention in the literature so far. As an aside, we clarify that for several well-known logics, their canonical characterization consequence operators are well-behaved.
An assistive embodied AI agent often has to collaborate with previously unseen humans. State of the art frameworks for such ad hoc teamwork use a large labeled dataset of prior observations to model the behavior of ot...
详细信息
ISBN:
(纸本)9783031742088;9783031742095
An assistive embodied AI agent often has to collaborate with previously unseen humans. State of the art frameworks for such ad hoc teamwork use a large labeled dataset of prior observations to model the behavior of other agents and to determine the ad hoc (i.e., embodied AI) agent's behavior. these approaches do not support rapid incremental revisions or transparency, and the necessary resources (e.g., training examples, computation) are not readily available in practical domains. Our previous work introduced an architecture that enabled an ad hoc agent to choose its actions in simple simulated domains based on non-monotonic logical reasoning with prior domain knowledge and models learned from limited examples to predict the behavior of other agents. Here, we extend this architecture to enable an ad hoc embodied AI agent to collaborate with a human performing household tasks in a complex indoor environment, focusing on the ad hoc agent's ability to identify and reason with relevant knowledge, and provide relational descriptions as explanations of its behavior and that of the human. We evaluate our architecture's capabilities in VirtualHome, a realistic 3D simulation environment.
In the fast-growing area of Artificial Intelligence (AI), the ability of autonomous agents to engage in complex debates is crucial for consensus building on beliefs, actions, or goals and forms the basis for applicati...
详细信息
ISBN:
(纸本)9783031742088;9783031742095
In the fast-growing area of Artificial Intelligence (AI), the ability of autonomous agents to engage in complex debates is crucial for consensus building on beliefs, actions, or goals and forms the basis for applications in decision-making, planning, opinion polling, and negotiation. In this paper, we leverage the Timed Concurrent Language for Argumentation, a modelling language derived from concurrent programming paradigms and Argumentation theory, to introduce well-known high-level propositions (claim, counter, why, argue, concede, and retract) to model various debate forms, making it a powerful tool for agent interaction. the obtained constructs, specifically designed for multi-agent reasoning and the facilitation of argumentation, define the dialogue language DICLA (DIalogic Concurrent Language for Argumentation) that enables domain experts to employ advanced computational argumentation tools without needing programming skills, bridging the gap between theoretical argumentation models and practical, real-world applications.
Finding non-conformities, such as physical failures causing electrical malfunctioning of a device, in modern semiconductor devices is challenging. Highly qualified employees in a failure analysis (FA) lab typically us...
详细信息
ISBN:
(纸本)9783031742088;9783031742095
Finding non-conformities, such as physical failures causing electrical malfunctioning of a device, in modern semiconductor devices is challenging. Highly qualified employees in a failure analysis (FA) lab typically use sophisticated and expensive tools like scanning electron microscopes to identify and locate such non-conformities. Given the increasing complexity of investigated devices and very limited resources, labs may struggle to deliver analysis results in time. this paper proposes an approach to optimize the usage of FA lab resources by combining constraint programming with stream reasoning enabling situation-dependent monitoring of the lab's conditions and schedule maintenance. Evaluation results indicate that our system can significantly improve the tardiness of real-world FA labs, and all its computational tasks can be finished in an average time of 3.6 s, with a maximum of 15.2 s, which is acceptable for the lab's workflows.
暂无评论