http://hdl.handle.net/10016/6652 2013-05-22T17:40:06Z 2013-05-22T17:40:06Z Moreno, Lourdes Martínez, Paloma Iglesias, Ana González, María http://hdl.handle.net/10016/16250 2013-02-12T14:42:21Z 2010-12-31T23:00:00Z

Title: HTML5 support for an accessible user-video-interaction on the Web Author(s): Moreno, Lourdes; Martínez, Paloma; Iglesias, Ana; González, María Abstract: Multimedia content covers the Web, and we should provide access to all people. For this reason, it is very important to take into account accessibility requirements in the player to avoid barriers and to ensure access to this multimedia content as well as their resources. One of the most frequent barriers is the technological obstacle: the necessity for the user to install the required plug-ins in to order to access video. The new standard HTML5 provides a solution to this problem. However, it does not fully support accessibility requirements of W3C standards, including WCAG and interaction requirement of UAAG. This paper introduces an overall study of this new standard in relation to accessibility requirements for the players as well as an accessible HTML5 Media Player.

2010-12-31T23:00:00Z Asensio, Javier Jiménez, Marta Fernández, Susana Borrajo, Daniel http://hdl.handle.net/10016/7728 2011-09-19T08:58:52Z 2007-12-31T23:00:00Z

Title: A social and emotional model for obtaining believable emergent behaviors Author(s): Asensio, Javier; Jiménez, Marta; Fernández, Susana; Borrajo, Daniel Abstract: This paper attempts to define an emotional model for virtual agents that behave autonomously in social worlds. We adopt shallow modeling based on the decomposition of the emotional state in two qualities: valence (pleasantness or hedonic value) and arousal (bodily activation) and, also, for the agent personality based on the five factors model (openness, conscientiousness, extroversion, agreeableness and neuroticism). The proposed model aims to endow agents with a satisfactory emotional state achieved through the social actions, i.e. the development of social abilities. Psychology characterizes these social abilities for: using the language as a tool (verbal and nonverbal communication), being learned, producing reciprocal reward among the individuals involved in the communication and for depending on the individual features. We have implemented our model in the framework of a computer game, AI-live, to show its validity. Description: Proceeding of: 13th International Conference, AIMSA 2008, Varna, Bulgaria, September 4-6, 2008

2007-12-31T23:00:00Z Fernández, Fernando Borrajo, Daniel http://hdl.handle.net/10016/7369 2011-09-19T08:30:27Z 1999-12-31T23:00:00Z

Title: VQQL. Applying vector quantization to reinforcement learning Author(s): Fernández, Fernando; Borrajo, Daniel Abstract: Reinforcement learning has proven to be a set of successful techniques for finding optimal policies on uncertain and/or dynamic domains, such as the RoboCup. One of the problems on using such techniques appears with large state and action spaces, as it is the case of input information coming from the Robosoccer simulator. In this paper, we describe a new mechanism for solving the states generalization problem in reinforcement learning algorithms. This clustering mechanism is based on the vector quantization technique for signal analog-to-digital conversion and compression, and on the Generalized Lloyd Algorithm for the design of vector quantizers. Furthermore, we present the VQQL model, that integrates Q-Learning as reinforcement learning technique and vector quantization as state generalization technique. We show some results on applying this model to learning the interception task skill for Robosoccer agents. Description: Proceeding of: RoboCup-99: Robot Soccer World Cup III, July 27 to August 6, 1999, Stockholm, Sweden

1999-12-31T23:00:00Z Giráldez, J. Ignacio Borrajo, Daniel http://hdl.handle.net/10016/6908 2012-11-15T12:44:17Z 1997-12-31T23:00:00Z

Title: Distributed reinforcement learning in multi-agent decision systems Author(s): Giráldez, J. Ignacio; Borrajo, Daniel Abstract: Decision problems can be usually solved using systems that implement different paradigms. These systems may be integrated into a single distributed system, with the expectation of obtaining a group performance more satisfactory than individual performances. Such a distributed system is what we call a Multi Agent Decision System (MADES), a special kind of Multi Agent System, that integrates several heterogeneous autonomous decision systems (agents). A MADES must produce a single solution proposal for the problem instance it faces, despite the fact that its decision making is distributed, and every agent produces solution proposals according to its local view and to its idiosyncrasy. We present a distributed reinforcement algorithm for learning how to combine the decisions the agents make in a distributed way, into a single group decision (solution proposal). Description: Proceeding of: 6th Ibero-American Conference on AI (IBERAMIA '98),Lisbon, Portugal, October 5–9, 1998

1997-12-31T23:00:00Z