Onaindia de la Rivaherrera, Eva

Organizational Units

Organizational Unit

Departamento de Sistemas Informáticos y Computación

Organizational Unit

Escuela Técnica Superior de Ingeniería Informática

Organizational Unit

Instituto Universitario Valenciano de Investigación en Inteligencia Artificial

Search Results

Now showing 1 - 10 of 40

Reactive execution for solving plan failures in planning control applications
(IOS Press, 2015-08-27) Gúzman Álvarez, César Augusto; Castejón Navarro, Pablo; Onaindia de la Rivaherrera, Eva; Frank, Jeremy; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial; Ministerio de Economía y Competitividad; Generalitat Valenciana
We present a novel reactive execution model for planning control applications which repairs plan failures at runtime. Our proposal is a domain-independent regression planning model which provides good-quality responses in a timely fashion. The use of a regressed model allows us to work exclusively with the sufficient and necessary information to deal with the plan failure. The model performs a time-bounded process that continuously operate on the plan to recover from incoming failures. This process guarantees there always exists a plan repair for a plan failure at anytime. The model is tested on a simulation of a real-world planetary space mission and on a well-known vehicle routing problem.
Mixed acceleration techniques for solving quickly stochastic shortest-path markov decision processes
(Universidad Nacional Autónoma de México (UNAM), 2011-08) García Hernández, Ma de Guadalupe; Ruiz Pinales, José; Onaindia de la Rivaherrera, Eva; Ledesma-Orozco, S; Aviña-Cervantes, J.G.; Alvarado-Méndez, E.; Reyes-Ballesteros, A; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial
In this paper we propose the combination of accelerated variants of value iteration mixed with improved prioritized sweeping for the fast solution of stochastic shortest-path Markov decision processes. Value iteration is a classical algorithm for solving Markov decision processes, but this algorithm and its variants are quite slow for solving considerably large problems. In order to improve the solution time, acceleration techniques such as asynchronous updates, prioritization and prioritized sweeping have been explored in this paper. A topological reordering algorithm was also compared with static reordering. Experimental results obtained on finite state and action-space stochastic shortest-path problems show that our approach achieves a considerable reduction in the solution time with respect to the tested variants of value iteration. For instance, the experiments showed in one test a reduction of 5.7 times with respect to value iteration with asynchronous updates.
Cooperative Multi-Agent Planning: A survey
(Association for Computing Machinery, 2017) Torreño Lerma, Alejandro; Onaindia de la Rivaherrera, Eva; Komenda, Antonín; tolba, Michal; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial; Generalitat Valenciana; Ministerio de Economía, Industria y Competitividad
[EN] Cooperative multi-agent planning (MAP) is a relatively recent research field that combines technologies, algorithms, and techniques developed by the Artificial Intelligence Planning and Multi-Agent Systems communities. While planning has been generally treated as a single-agent task, MAP generalizes this concept by considering multiple intelligent agents that work cooperatively to develop a course of action that satisfies the goals of the group. This article reviews the most relevant approaches to MAP, putting the focus on the solvers that took part in the 2015 Competition of Distributed and Multi-Agent Planning, and classifies them according to their key features and relative performance.
Plan commitment: Replanning versus plan repair
(Elsevier, 2023-08) Babli, Mohannad; Sapena Vercher, Oscar; Onaindia de la Rivaherrera, Eva; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial
[EN] While executing its plan in a dynamic environment where multiple agents are operating, an autonomous agent may suffer a failure due to discrepancies between the expected and actual context and thus must replace its obsolete plan. In its endeavour to fix the failure and reach its original goals, the agent may unknowingly disrupt other agents executing their plans in the same environment. We present a property for plan repair called plan commitment to ensure a responsible repair policy among agents that aims to minimise the negative impact on others. We present arguments to support the claim that plan commitment is a valuable property when an agent may have made bookings and commitments to others. We then propose C-TFLAP, an implementation of a plan repair heuristic that allows adapting a failed plan to the new context while committing as much as possible to the original plan. We demonstrate empirically that: (1) our plan repair achieves more committed plans than plan-stability repair when an agent has made bookings and commitments to others, and (2) compared to typical replanning and plan-stability repair, it can reduce the revisions among agents when failures are avoidable and can decrease the time-loss otherwise. In addition, to demonstrate extensibility, we integrate context-aware knowledge extension with committed repairing to increase the agent¿s chances of repairing.
A Decentralized Multi-Agent Coordination Method for Dynamic and Constrained Production Planning
(International Foundation for Autonomous Agents and Multiagent Systems, 2020-05-13) Lujak, Marin; Fernandez, Alberto; Onaindia de la Rivaherrera, Eva; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial; Agencia Estatal de Investigación; European Regional Development Fund; European Cooperation in Science and Technology
[EN] In the capacitated production planning problem, quantities of products need to be determined at consecutive periods within a given time horizon when product demands, costs, and production capacities vary through time. We focus on a general formulation of this problem where each product is produced in one step and setup cost is paid at each period of production. Additionally, products can be anticipated or backordered in respect to the demand period. We propose a computationally efficient decentralized approach based on the spillover effect relating to the accumulation of production costs of each product demand through time. The performance of the spillover algorithm is compared against the state-of-the-art mixed integer programming branch-and-bound solver CPLEX 12.8 considering optimality gap and computational time.
A common framework for learning causality
(Springer-Verlag, 2018) Onaindia de la Rivaherrera, Eva; Aineto, Diego; Jiménez Celorrio, Sergio; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial; Ministerio de Educación, Cultura y Deporte; Agencia Estatal de Investigación; Ministerio de Economía y Competitividad
[EN] Causality is a fundamental part of reasoning to model the physics of an application domain, to understand the behaviour of an agent or to identify the relationship between two entities. Causality occurs when an action is taken and may also occur when two happenings come undeniably together. The study of causal inference aims at uncovering causal dependencies among observed data and to come up with automated methods to find such dependencies. While there exist a broad range of principles and approaches involved in causal inference, in this position paper we argue that it is possible to unify different causality views under a common framework of symbolic learning.
Special issue on agreement technologies
(Springer Verlag (Germany), 2015-08) Ivan Chesnevar, Carlos; Onaindia de la Rivaherrera, Eva; Ossowski, Sascha; Vouros, George; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial
Most would agree that large-scale open distributed systems are an area of enormous social and economic potential. In fact, regardless of whether they are realised for the purpose of business or leisure, in a private or a public context, people’s transactions and interactions are increasingly mediated by computers. The resulting networks are usually large in scale, involving millions of interactions, and are open for the interacting entities to join or leave at will. We have been experiencing a paradigm shift in the way that such systems are built, enacted, and managed: away from rigid and centralised client–server architectures, towards more flexible and decentralised means of interaction. Nowadays, there is a growing awareness that the notion of agreement and agreement processes will be of key importance for the next generation of such decentralised large-scale open distributed systems (Ossowski et al. 2013).
Can Tourist Attractions Boost Other Activities Around? A Data Analysis through Social Networks
(MDPI AG, 2019-06-01) Bustamante, Alexander; Sebastiá Tarín, Laura; Onaindia de la Rivaherrera, Eva; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Geodésica, Cartográfica y Topográfica; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial; Agencia Estatal de Investigación
[EN] Promoting a tourist destination requires uncovering travel patterns and destination choices, identifying the profile of visitors and analyzing attitudes and preferences of visitors for the city. To this end, tourism-related data are an invaluable asset to understand tourism behaviour, obtain statistical records and support decision-making for business around tourism. In this work, we study the behaviour of tourists visiting top attractions of a city in relation to the tourist influx to restaurants around the attractions. We propose to undertake this analysis by retrieving information posted by visitors in a social network and using an open access map service to locate the tweets in a influence area of the city. Additionally, we present a pattern recognition based technique to differentiate visitors and locals from the collected data from the social network. We apply our study to the city of Valencia in Spain and Berlin in Germany. The results show that, while in Valencia the most frequented restaurants are located near top attractions of the city, in Berlin, it is usually the case that the most visited restaurants are far away from the relevant attractions of the city. The conclusions from this study can be very insightful for destination marketers.
Temporal landmark graphs for solving overconstrained planning problems
(Elsevier, 2016-08-15) Marzal Calatayud, Eliseo Jorge; Sebastiá Tarín, Laura; Onaindia de la Rivaherrera, Eva; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Geodésica, Cartográfica y Topográfica; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial; Ministerio de Economía y Competitividad
This paper presents TempLM, a novel approach for handling temporal planning problems with deadlines. The proposal revolves around the concept of temporal landmark, a proposition that must be necessarily true in all solution plans to achieve the problem goals within their deadlines. The temporal landmarks extracted from the problem form a landmarks graph where nodes are landmarks and edges represent temporal as well as causal relationships between landmarks. The graph comprises information about which propositions and when these propositions must be achieved in a solution plan, information that is later used to guide the search process as well as reduce the search space. Thus, the partial plans of the search tree that are not compliant with the information comprised in this graph are pruned. We present an exhaustive experimentation evaluation in overconstrained and unsolvable problems and we compare the performance of TempLM with other state-of-the-art planners. The results will show the efficiency of TempLM in the detection of unsolvable problems. (C) 2016 Elsevier B.V. All rights reserved:
Automated feature extraction for planning state representation
(Sociedad Iberoamericana de Inteligencia Artificial (IBERAMIA), 2024-12) Sapena Vercher, Oscar; Onaindia de la Rivaherrera, Eva; Marzal Calatayud, Eliseo Jorge; Departamento de Sistemas Informáticos y Computación; Escuela Técnica Superior de Ingeniería Informática; Instituto Universitario Valenciano de Investigación en Inteligencia Artificial; Agencia Estatal de Investigación; European Regional Development Fund
[EN] Deep learning methods have recently emerged as a mechanism for generating embeddings of planning states without the need to predefine feature spaces. In this work, we advocate for an automated, cost-effective and interpretable approach to extract representative features of planning states from high-level language. We present a technique that builds up on the objects type and yields a generalization over an entire planning domain, enabling to encode numerical state and goal information of individual planning tasks. The proposed representation is then evaluated in a task for learning heuristic functions for particular domains. A comparative analysis with one of the best current sequential planner and a recent ML-based approach demonstrate the efficacy of our method in improving planner performance. Resumen Los meto dos de aprendizaje profundo han surgido recientemente como un mecanismo para generar embeddings de estados de planificacion sin la necesidad de predefinir espacios de caracter & imath;sticas. En este trabajo, abogamos por un enfo que automatizado, eficiente en costes e interpretable para extraer caracter& imath;sticas representativas de los estados de planificacion a partir de un lenguaje de alto nivel. Presentamos una tecnica que se basa en los tipos de objetos y permite una generalizacion sobre todo un dominio de planificacion, posibilitando la codificacion de informacion numerica del estado y de los objetivos de tareas de planificacion individuales. La representacion propuesta se evalua mediante una tarea de aprendizaje de funciones heur& imath;sticas para dominios espec& imath;ficos. Un analisis comparativo con uno de los mejores planificadores secuenciales actuales y con un enfo que reciente basado en aprendizaje automatico demuestra la eficacia de nuestro metodo para mejorar el rendimiento de los planificadores.