Dr. Raúl Arrabales MorenoList of Publications
 Arrabales, R. "Evaluation and Development of Consciousness in Artificial Cognitive Systems". PhD Thesis. Carlos III University of Madrid. February, 2011.
[PDF] [PPT] [BibTeX] Abstract
Historically human consciousness has been rather excluded from the scientific debates, being philosphy the most important perspective for its study. However, over the last three decades different research disciplines such as philosophy of mind and cognitive psychology have shown a growing interest on the problem of consciousness. This trend has taken place also in the multidisciplinary context of neuroscience. In fact, recent advances in neuroimaging techniques have led most researchers to consider consciousness as a subject for scientific study. The development of new biological and psychological theories on the production of consciousness in humans has revived the original challenges of Artificial Intelligence. Specifically, the field of Machine Consciousness is becoming a rigorous scientific discipline aimed at studying and potentially creating machines with different types and levels of consciousness. In the context of Machine Consciousness, this thesis aims at contributing to the scientific knowledge about consciousness by means of two interrelated research lines: the first one consists of the conception and application of a novel method for the measurement and characterization of the level of development of consciousness of an artificial agent; the second one is based on an artificial cognitive architecture inspired on several theories of consciousness. The application of the proposed measuring technique will permit the detailed analysis of the current level of development of conscious machines and to identify what are the aspects that have not yet been achieved. Furthermore, the proposed scale will be used as a roadmap to identify what are the key cognitive skills that need to be implemented in order to create human-like machines. The application of the proposed cognitive architecture as a fundamental component of the control system of different artificial agents will permit the experimentation with different cognitive functions associated to consciousness. The interaction between different capabilities like attention, emotions or sensory prediction will be analyzed, looking for potential synergies that produce complex and adaptive behaviors. Additionally, using the proposed computational model of consciousness, a synthetic phenomenology approach will be adopted based on the modeling of the contents of conscious experience. The conscious experience reported by a human subject when confronted to certain stimuli will be compared with the explicit content that the cognitive architecture is able to generate when confronted to the same stimuli. The results obtained from these research lines will provide valuable information about the validity of the theories of consciousness that have been analyzed as well as the differences between natural and artificial cognitive processes. Besides, possible areas of application of the proposed cognitive architecture will be explored, such us the creation of artificial agents able to develop believable human-like behaviors. Arrabales, R. Ledezma, A. and Sanchis, A. "ConsScale FPS: Cognitive Integration for Improved Believability in Computer Game Bots". In Believable Bots. Hingston. P. (ed). Springer, 2012. pp. 193-214. ISBN 978-3-642-32322-5.
[BOOK] [Chapter] [BibTeX] Abstract ConsScale is a framework for the characterization and measurement of the level of cognitive development in artificial agents. The scale is inspired by an evolutionary perspective of the development of consciousness in biological organisms. In this context, consciousness is considered as an evolutionary advantage that provides the individual with a highly adaptive super-function, i.e. effective synergistic integration of a number of cognitive functions. ConsScale FPS is a specific instantiation of the scale aimed at describing the hierarchical structure of the cognitive skills required by a bot in the domain of a first-person shooter (PFS) videogame. Using ConsScale FPS as reference framework, this work proposes specific strategies for the design of more believable bots. The concept of cognitive dependency is introduced and cognitive abilities implemented in state of the art bots are analyzed from the point of view of effective cognitive integration. Finally, in light of the cognitive development of current bots, particular improvements are suggested towards the design of more human-like bots.
Arrabales, R. "El mundo interior de Peca". Diego Sampredro Ruiz (Coord). Un Breve Viaje por la Ciencia. Universidad de La Rioja. 2012. pp. 33-38. ISBN 978-84-96487-69-7.
[BOOK] [Chapter] Abstract Como buen científico no dejo de hacerme preguntas sobre la realidad que nos rodea. Las preguntas que a menudo me parecen más interesantes están relacionadas con las fronteras de la propia ciencia. ¿Puede proporcionarnos la ciencia la verdad acerca de todas las cosas? ¿Dónde está el límite? ¿Es aplicable el método científico a todos los fenómenos conocidos?
Arrabales, R., Muñoz, J. et. al. "A Machine Consciousness Approach to the Design of human-like bots". In Believable Bots. Hingston. P. (ed). Springer, 2012. pp. 171-191. ISBN 978-3-642-32322-5.
[BOOK] [Chapter] [BibTeX] Abstract This chapter introduces Machine Consciousness as a new research field applied to the development of non-player characters (NPC) human-like behaviour in video games. Key aspects, advantages, and challenges of this young research area are discussed using the cognitive architecture CERA-CRANIUM as an illustrative example of an autonomous control system inspired by cognitive theories of human consciousness. Additionally, other cognitive architectures used in video games are also analyzed. The bot codenamed CC-Bot2, winner of the 2K BotPrize 2010 competition and based on the CERA-CRANIUM cognitive architecture is also described in this chapter. Specifically, the particular way in which CC-Bot2 processes the sensorimotor information and generates sequences of adaptive human-like actions is discussed. We also analyze the main differences between CC-Bot2 and other bots, focusing on the key features that allowed CC-Bot2 to reach the first place in the competition. Finally, we conclude describing the main lines of work for future CCBot implementations and pointing out major conclusions about the application of Machine Consciousness to the design of believable bots.
Arrabales, R. and Muñoz, J. "Enhancing Game Play Experience Using Machine Consciousness". Video Game Play and Consciousness. In Series: Perspectives on Cognitive Psychology, Computer Science, Technology and Applications, J. Gackenbach (Ed.). Nova. pp. 319-328. ISBN: 978-1-62081-113-9. [BOOK] [BibTeX] Abstract Machine Consciousness is a young cross-disciplinary field of research primarily aimed at both understanding what is consciousness and how could it be simulated or replicated in artificial agents. One of the possible application domains for Machine Consciousness implementations is the design of human-like video game’s non-player characters (NPC). Generating human-like behavior in NPCs is interesting from the point of view of game play experience, however it is very challenging. Usually, it is straightforward to design artificial characters with very simple and limited behavior, and it is even possible to create very good artificial players that outweigh typical human playing skills. However, generating the characteristic engaging human-like behavior is much more problematic and remains an open problem. In this chapter, the use of an artificial control architecture based on cognitive theories of consciousness is described and its application to the generation of more appealing and believable human-like behavior in a video game NPC is discussed. Specifically, it is argued that the use of Machine Consciousness techniques for the design and implementation of NPCs can improve the experience of the player.
Arrabales, R. Ledezma, A. and Sanchis, A. "Simulating Visual Qualia in the CERA-CRANIUM Cognitive Architecture". From Brains to Systems: Brain-Inspired Cognitive Systems 2010. In Series: Advances in Experimental Medicine and Biology, vol. 718, Hernández, C.; Sanz, R.; Gomez, J.; Smith, L.S.; Hussain, A.; Chella, A.; Aleksander, I. (Eds.). Springer New York. In press. ISBN: 978-1-4614-0163-6.
[PDF] [BOOK] [BibTeX] Abstract The concept of qualia poses a central problem in the framework of consciousness studies. Despite it being a controversial issue even in the study of human consciousness, we argue that qualia can be complementarily studied using artificial cognitive architectures. In this work we address the problem of defining qualia in the domain of artificial systems, providing a model of “artificial qualia”. Furthermore, we partially apply the proposed model to the generation of visual qualia using the cognitive architecture CERA-CRANIUM, which is modeled after the global workspace theory of consciousness. It is our aim to define, characterize and identify artificial qualia as direct products of a simulated conscious perception process. Simple forms of the apparent motion effect are used as the basis for a preliminary experimental setting focused on the simulation and analysis of synthetic visual experience. In contrast with the study of biological brains, the inspection of the dynamics and transient inner states of the artificial cognitive architecture can be performed effectively, thus enabling the detailed analysis of covert and overt percepts generated by the system when it is confronted with specific visual stimuli. The observed states in the artificial cognitive architecture during the simulation of apparent motion effects are used to discuss the existence of possible analogous mechanisms in human cognition processes. | JCR Indexed Journals (2): |
Arrabales, R. Ledezma, A. and Sanchis, A. "ConsScale: A Pragmatic Scale for Measuring the Level of Consciousness in Artificial Agents". Journal of Consciousness Studies. Vol. 17. No. 3-4. March-April 2010. pp. 131-164(34)
[JCR Social Sciences, Interdisciplinary (IF:0.918)(Quartile Q2)(Top Third)]
[PDF] [ConsScale] [BibTeX] [Journal Website] Abstract
One of the key problems the field of Machine Consciousness (MC) is currently facing is the need to accurately assess the potential level of consciousness that an artificial agent might develop. This paper presents a novel artificial consciousness scale designed to provide a pragmatic and intuitive reference in the evaluation of MC implementations. The version of ConsScale described in this work provides a comprehensive evaluation mechanism which enables the estimation of the potential degree of consciousness of most of the existing artificial implementations. This scale offers both well defined levels of artificial consciousness (that can be used for qualitative classification of agents) and a method to calculate an orientative numerical score (which provides a quantitative grade for comparing agents in terms of consciousness). A set of architectural and cognitive criteria is considered for each level of the scale. This permits the definition of a cognitive framework in which MC implementations can be ranked according to their potential capability to reproduce functional synergies associated with consciousness. The probability of the implementations having any phenomenal states related to the assessed functional synergy is not specifically addressed in this paper; nevertheless, it could be thoughtfully discussed elsewhere.
Arrabales, R. and Sanchis de Miguel, A. "Applying Machine Consciousness Models in Autonomous Situated Agents". Pattern Recognition Letters. Special Issue on Pattern Recognition in Multidisciplinary Perception and Intelligence. Volume 29. Issue 8. Pages 1033-1038. June 2008.
[JCR Computer Science, Artificial Intelligence (IF: 1.559)(Quartile Q2)(Second Third)]
[PDF] [BibTeX] [DOI] [Journal Website] Abstract
This paper briefly describes the most relevant current approaches to the implementation of scientific models of consciousness. Main aspects of scientific theories of consciousness are characterized in sight of their possible mapping into artificial implementations. These implementations are analyzed both theoretically and functionally. Also, a novel pragmatic functional approach to machine consciousness is proposed and discussed. A set of axioms for the presence of consciousness in agents is applied to evaluate and compare the various models. | Peer Reviewed International Journals (5): |
Arrabales, R. "Inner Speech Generation in a Video Game Non-Player Character: From Explanation to Self?". International Journal of Machine Consciousness. Vol. 4. No. 2. pp. 367-381. December 2012.
[PDF] [DOI] [BibTeX] [Journal Website] Abstract
The use of human language is a hallmark of human consciousness, even when it is not used publicly. Inner speech is the way humans consciously communicate with themselves and arguably a key factor contributing to the formation of more self-aware selves. From the perspective of cognitive science and artificial cognitive architectures, inner speech can be also seen as a metamanagement system that modulates some cognitive processes of the subject. In this paper, we describe a preliminary version of a computational model of inner speech generation based on the cognitive architecture CERA-CRANIUM. This inner speech generation method is illustrated using a video game non-player character as the subject of the first-person narratives to be produced. We also use this model of inner speech generation to discuss the possibilities of using such a first-person narrative stream as a meta-control input to the artificial cognitive architecture. We argue that this verbal input might be used as an integrated self-explanation of the agent in the world and thus contribute to the formation of self.
Arrabales, R. Ledezma, A. and Sanchis, A. "The Cognitive Development of Machine Consciousness Implementations". International Journal of Machine Consciousness. Vol 2. No. 2. Alexei V. Samsonovich (Eds), pp. 213-225. December 2010. [PDF] [DOI] [BibTeX] [Journal Website] Abstract
The progress in the machine consciousness research field has to be assessed in terms of the features demonstrated by the new models and implementations currently being designed. In this paper, we focus on the functional aspects of consciousness and propose the application of a revision of ConsScale – a biologically inspired scale for measuring cognitive development in artificial agents – in order to assess the cognitive capabilities of machine consciousness implementations. We argue that the progress in the implementation of consciousness in artificial agents can be assessed by looking at how key cognitive abilities associated to consciousness are integrated within artificial systems. Specifically, we characterize ConsScale as a partially ordered set and propose a particular dependency hierarchy for cognitive skills. Associated to that hierarchy a graphical representation of the cognitive profile of an artificial agent is presented as a helpful analytic tool. The proposed evaluation schema is discussed and applied to a number of significant machine consciousness models and implementations. Finally, the possibility of generating qualia and phenomenological states in machines is discussed in the context of the proposed analysis.
Arrabales, R. Ledezma, A. and Sanchis, A. "Strategies for Measuring Machine Consciousness". International Journal of Machine Consciousness. Vol 1. Issue 2. December 2009. pp. 193-201.
[PDF] [BibTeX] [DOI] [Journal Website] Abstract
The accurate detection and measurement of the level of consciousness of a creature remains a major scientific challenge, nevertheless a number of new accounts that attempt to address this problem have been proposed recently. In this paper we analyze the principles of these new measures of consciousness along with other classical approaches focusing on their applicability to Machine Consciousness (MC). Furthermore, we propose a set of requirements of what we think a suitable measure for MC should be, discussing the associated theoretical and practical issues. Using the proposed requirements as a framework for the design of an integrative measure of consciousness, we explore the possibility of designing such a measure in the context of current state of the art of consciousness studies. We conclude this analysis with a review of the major problems that still need to be addressed if the proposed strategy is followed.
Arrabales, R. Ledezma, A. and Sanchis, A. "A Cognitive Approach to Multimodal Attention". Journal of Physical Agents. Volume 3. Issue 1. Pages 53-64. January 2009.
[PDF] [BibTeX] [Journal Website] Abstract
An adaptive attention mechanism is a requirement when an autonomous robot has to deal with real world environments. In this paper we present a novel cognitive architecture which enables integrated and efficient filtering of multiple modality sensory information. The proposed attention mechanism is based on contexts that determine what sensorimotor data is relevant to the current situation. These contexts are used as a mean to adaptively select constrained cognitive focus within the vast multimodal sensory space. In this framework, the focus of attention can be directed to meaningful complex percepts, thus allowing the implementation of higher cognitive capabilities. Sonar, contact, and visual sensory modalities have been used in the perception process, and the motor capability of the physical agent is provided by a differential wheel drive system. The testing of this artificial attention approach, carried out initially in the domain of counterpart recognition and chasing, has demonstrated both a great decrease in computation power requirements and ease of multimodal integration for cognitive representations.
Arrabales, R. Ledezma, A. and Sanchis, A. "Integrated Attention for Cognitive Robotics". Communications of the Systemics and Informatics World Network (SIWN). Volume 5. August 2008. Pages 1-5.
[PDF] [BibTeX] [Journal Website] Abstract
An autonomous robot is usually expected to perform its assigned task in complex and unstructured environments, where the high amount of sensory information acquired by sensors cannot be entirely processed in real-time. Like in living organisms, an attention mechanism is required in order to select most relevant stimuli from the environment. The outcome of this selection process comprises all the information that will be available to the control system of the robot. Therefore, an optimal operation of the attention mechanism is necessary for robot mission success. In this work, we present a novel cognitive architecture which permits the implementation and integration of efficient attention mechanisms for autonomous robots. Additionally, the application of this approach to the domain of unknown environment exploration is discussed and results are analyzed. Specifically, the influence of attention in the generation of efficient robot behavior is demonstrated. | International Conference Proceedings (9): |
Arrabales, R. Ledezma, A. and Sanchis, A. "Towards the Generation of Visual Qualia in Artificial Cognitive Architectures". BICS 2010 (Brain Inspired Cognitive Systems). Conference Preprints ISBN 978-84-614-1870-1.
[PDF] [BibTeX] [Conference Website] Abstract
The nature and the generation of qualia in machines is a highly controversial issue. Even the existence of such a concept in the realm of artificial systems is often neglected or denied. In this work, we adopt a pragmatic approach to this problem using the Synthetic Phenomenology perspective. Specifically, we explore the generation of visual qualia in an artificial cognitive architecture inspired on the Global Workspace Theory (GWT). We argue that preliminary results obtained as part of this research line will help to characterize and identify artificial qualia as the direct products of conscious perception in machines. Additionally, we provide a computational model for integrated covert and overt perception in the framework of the GWT. A simple form of the apparent motion effect is used as a preliminary experimental context and a practical case study for the generation of synthetic visual experience. Thanks to an internal inspection subsystem, we are able to analyze both covert and overt percepts generated by our system when confronted with visual stimuli. The inspection of the internal states generated within the cognitive architecture enable us to discuss possible analogies with human cognition processes.
Arrabales, R. Ledezma, A. and Sanchis, A. "On the Practical Nature of Artificial Qualia". Symposium on AI-Inspired Biology. The 2010 Annual Convention of the Society for the Study of Artificial Intelligence and Simulation of Behaviour (AISB2010). Leicester, UK. March 2010. In press.
[PDF] [BibTeX] [Symposium Website] Abstract
Can machines ever have qualia? Can we build robots with inner worlds of subjective experience? Will qualia experienced by robots be comparable to subjective human experience? Is the young field of Machine Consciousness (MC) ready to answer these questions? In this paper, rather than trying to answer these questions directly, we argue that a formal definition, or at least a functional characterization, of artificial qualia is required in order to establish valid engineering principles for synthetic phenomenology (SP). Understanding what might be the differences, if any, between natural and artificial qualia is one of the first questions to be answered. Furthermore, if an interim and less ambitious definition of artificial qualia can be outlined, the corresponding model can be implemented and used to shed some light on the very nature of consciousness. In this work we explore current trends in MC and SP from the perspective of artificial qualia, attempting to identify key features that could contribute to a practical characterization of this concept. We focus specifically on potential implementations of artificial qualia as a means to provide a new interdisciplinary tool for research on natural and artificial cognition.
Arrabales, R. Ledezma, A. and Sanchis, A. "Assessing and Characterizing the Cognitive Power of Machine Consciousness Implementations". AAAI 2009 Fall Symposium on Biologically Inspired Cognitive Architectures (BICA-2009). Technical Report FS-09-01. pp. 16-21. AAAI Press. ISBN 978-1-57735-435-2.
[PDF] [AAAI] [PPT] [APP] [BibTeX] [Conference Website] [Proceedings] [Pics] Abstract
Many aspects can be taken into account in order to assess the power and potential of a cognitive architecture. In this paper we argue that ConsScale, a cognitive scale inspired on the development of consciousness, can be used to characterize and evaluate cognitive architectures from the point of view of the effective integration of their cognitive functionalities. Additionally, a graphical characterization of the cognitive power of artificial agents is proposed as a helpful tool for the analysis and comparison of Machine Consciousness implementations. This is illustrated with the application of the scale to a particular problem domain in the context of video game synthetic bots. Also, the problem of defining general architectural and behavioral criteria and their relation with the possibility of machine phenomenal states is briefly discussed.
Arrabales, R. Ledezma, A. and Sanchis, A. "Towards Conscious-like Behavior in Computer Game Characters", in Proceedings of the IEEE Symposium on Computational Intelligence and Games 2009 (CIG-2009) pp. 217-224. ISBN 978-1-4244-4815-9.
[CORE ERA Conference Ranking: C]
[PDF] [PPT] [DOI] [BibTeX] [Conference Website] Abstract
The main sources of inspiration for the design of more engaging synthetic characters are existing psychological models of human cognition. Usually, these models, and the associated Artificial Intelligence (AI) techniques, are based on partial aspects of the real complex systems involved in the generation of human-like behavior. Emotions, planning, learning, user modeling, set shifting, and attention mechanisms are some remarkable examples of features typically considered in isolation within classical AI control models. Artificial cognitive architectures aim at integrating many of these aspects together into effective control systems. However, the design of this sort of architectures is not straightforward. In this paper, we argue that current research efforts in the young field of Artificial Consciousness (AC) could contribute to tackle complexity and provide a useful framework for the design of more appealing synthetic characters. This hypothesis is illustrated with the application of a novel consciousness-based cognitive architecture to the development of a First Person Shooter video game character.
Arrabales, R. Ledezma, A. and Sanchis, A. "Establishing a Roadmap and Metrics for Conscious Machines Development". In Proceedings of the 8th IEEE International Conference on Cognitive Informatics. G. Baciu, Y. Wang, Y.Y. Yao, W. Kinsner, K. Chan & L.A. Zadeh (Eds.). Hong Kong. June 2009. Pages 94-101. ISBN 978-1-4244-4642-1.
[CORE ERA Conference Ranking: C]
[PDF] [PPT] [DOI] [BibTeX] [Conference Website] Abstract
From the point of view of Cognitive Informatics, consciousness can be considered as a grand integration of a number of cognitive processes. Intuitive definitions of consciousness generally involve perception, emotions, attention, self-recognition, theory of mind, volition, etc. Due to this compositional definition of the term consciousness it is usually difficult to define both what is exactly a conscious being and how consciousness could be implemented in artificial machines. When we look into the most evolved biological examples of conscious beings, like great apes or humans, the vast complexity of observed cognitive interactions in conjunction with the lack of comprehensive understanding of low level neural mechanisms makes the reverse engineering task virtually unreachable.
With the aim to effectively address the problem of modeling consciousness at a cognitive level, in this work we propose a concrete developmental path in which key stages in the progressive process of building conscious machines are identified and characterized. Furthermore, a method for calculating a quantitative measure of artificial consciousness is presented. The application of the proposed framework is illustrated with the comparative study of different software agents designed to compete in a first-person shooter video game.
Arrabales, R. Ledezma, A. and Sanchis, A. "Designing Human-like Video Game Synthetic Characters through Machine Consciousness". Towards a Science of Consciousness 2009. Hong Kong. June 2009. Pages 74.
[PDF] [PPT] [BibTeX] [Conference Website] Abstract
In recent years we are witnessing a huge growth of the video game industry. One of the factors that have contributed to current technology in video games is the improvement of Artificial Intelligence bots (AI bots for short), which are software agents able to autonomously interact with other AI bots and human players. However, the latest great figures achieved by the video game industry are actually due to the proliferation of online games. Why online gaming is so appealing to players? We believe one of the main reasons is the intelligence and consciousness showed by opponents in these scenarios. In a multiplayer online game, the player competes and collaborates more or less directly with other humans, hence obtaining an engaging experience. Although current AI bots can be intelligent to some extend, they cannot provide the same level of conscious-like behavior as produced by a human player [...]
Arrabales, R. Ledezma Espino, A. and Sanchis de Miguel, A. "Modelling Consciousness for Autonomous Robot Exploration". In 2nd International Work-Conference on the Interplay between Natural and Artificial Computation, IWINAC 2007. Lecture Notes in Computer Science Series, Vol. 4527. pp. 51-60.
[PDF] [BibTeX] [RefWorks] [ProCite/EndNote] [Conference] Abstract
This work aims to describe the application of a novel machine consciousness model to a particular problem of unknown environment exploration. This relatively simple problem is analyzed from the point of view of the possible benefits that cognitive capabilities like attention, environment awareness and emotional learning can offer. The model we have developed integrates these concepts into a situated agent control framework, whose first version is being tested in an advanced robotics simulator. The implementation of the relationships and synergies between the different cognitive functionalities of consciousness in the domain of autonomous robotics is also discussed.
Arrabales, R. and Sanchis de Miguel, A., "La Aplicación de Modelos de Consciencia Artificial en los Sistemas Multiagente". Actas del Campus Multidisciplinar en Percepción e Inteligencia, CMPI-2006, Vol. 1 pp. 401-412. July 2006. ISBN 84-689-9560-6.
[Congress] Abstract
Durante la última década han aparecido algunas implementaciones de modelos científicos de la consciencia basados en sistemas multiagente. El propósito de este artículo es recopilar y describir estos sistemas, determinando hasta que punto estas implementaciones satisfacen los modelos correspondientes, y analizando si proporcionan realmente las supuestas ventajas de usar consciencia artificial en la resolución de problemas. También se analizan en general las funciones de la consciencia y los beneficios que éstas pueden aportar en el rendimiento de los sistemas multiagente.
Arrabales R., Flanagan C., Toal D., "An adaptive video event mining system for an autonomous underwater vehicle" In: Intelligent Engineering Systems through Artificial Neural Networks, 2002, Vol 12, pp 585-591, ASME Press, Proc. ANNIE 2002, St Louis. ISBN 0-7918-0191-8.
[CORE ERA Conference Ranking: B]
[PDF] [BibTeX] [RefWorks] [ProCite/EndNote] [Book] Abstract
This paper presents the results obtained in the development of an adaptive architecture for automated event discovery in sub-sea recorded video footage. The Video Marking System that has been built is the first step of the development of a vision system for an autonomous underwater vehicle. The principal aim of our work is to build an adaptive architecture that provides the vision system with the intelligence and robustness required to deal with the great variability found in underwater video. Different image processing techniques are embedded within the adaptive architecture; all of them running in parallel, adaptively parameterized, and assigned a confidence level by a voting system. Utilizing a simplified world model, the various outputs from the active image processing approaches are combined into an improved event description. | International Workshop Proceedings (6): |
Arrabales, R. "Exploring Possible Mechanisms for Inner Speech Generation in the CERA-CRANIUM Cognitive Architecture". Retecog 2011 - The Architectures of Mind. Madrid. July 2011.
[PDF] [Workshop Website] Abstract
Experiencing visual imagery and having inner dialogues are probably the most familiar hallmarks of human consciousness. We argue that these key aspects of consciousness can be studied from the point of view of Machine Consciousness considering them as high-level cognitive functions. Consequently, we believe that using artificial cognitive architectures the processes associated with these functions could be designed, implemented, simulated, and analyzed. In this paper we focus specifically on the generation of inner speech using the CERA-CRANIUM cognitive architecture [1]. As CERA-CRANIUM is based on a computational model of the Global Workspace Theory (GWT) [3], we explore the most plausible mechanisms that might support the generation of inner speech in the context of the GWT. Based on our former work on the generation of visual qualia using the CERACRANIUM computational model [2], here we aim at characterizing inner speech as the result of a complex process of self-narration of both the current explicit perception flow of the system and the decision making mechanism. Therefore, the input of such process would be the sequential thread of explicit percepts autonomously generated by the cognitive architecture and the sequence of interim states of the decision making subsystem. An interesting key aspect of this model is that the role of inner speech in decision making is explained by the feedback loop formed by “inner speech percepts” that become part of the thread of explicit percepts of the agent (conscious content of the
artificial mind which is used for explicit reasoning). Using the former hypothesis, we discuss the functional advantages of having a mechanism for inner speech in an artificial agent. Additionally we propose specific computational models for a practical implementation and present the results of preliminary experiments using a CERA-CRANIUM controlled agent in a video game.
Arrabales, R. Ledezma, A. and Sanchis, A. "CERA-CRANIUM: A Test Bed for Machine Consciousness Research". International Workshop on Machine Consciousness 2009. Hong Kong. June 2009. Pages 105.
[PDF] [PPT] [BibTeX] [Workshop Website] Abstract
This paper describes a novel framework designed as a test bed for machine consciousness cognitive models (MCCM). This MCCM experimentation framework is based on a general-purpose cognitive architecture that can be integrated in different environments and confronted with different problem domains. The definition of a generic cognitive control system for abstract agents is the root of the versatility of the presented framework. The proposed control system, which is inspired in the major cognitive theories of consciousness, provides mechanisms for both sensory data acquisition and motor action execution. Sensory and motor data is represented in the proposed architecture using different level workspaces where percepts and actions are generated thanks to the competition and collaboration of specialized processors. Additionally, this cognitive architecture provides the means to modulate perception and behavior; in other words, it offers an interface for a higher control layer to drive the way percepts and actions are generated and how they interact with each other. This mechanism permits the experimentation with virtually any high level cognitive model of consciousness. An illustrative application scenario, autonomous explorer robots, is also reviewed in this work.
Arrabales, R. Ledezma, A. and Sanchis, A. "ConsScale: A Plausible Test for Machine Consciouness?". Proceedings of the Nokia Workshop on Machine Consciousness - 13th Finnish Artificial Intelligence Conference (STeP 2008). Helsinki. Finland. Pages 49-57.
[PDF] [PPT] [BibTeX] [Conference Website] [Pics] Abstract
Is consciousness a binary on/off property? Or is it on the contrary a complex phenomenon that can be present in different states, qualities, and degrees? We support the latter and propose a linear incremental scale for consciousness applicable to artificial agents. ConsScale is a novel agent taxonomy intended to classify agents according to their level of consciousness. Even though testing for consciousness remains an open question in the domain of biological organisms, a review of current biological approaches is discussed as well as their possible adapted application into the realm of artificial agents.
Regarding to the always controversial problem of phenomenology, in this work we have adopted a purely functional approach, in which we have defined a set of architectural and behavioral criteria for each level of consciousness. Thanks to this functional definition of the levels, we aim to specify a set of tests that can be used to unambiguously determine the higher level of consciousness present in the artificial agent under study. Additionally, since a number of objections can be presumably posed against our proposal, we have considered the most obvious critiques and tried to offer reasonable rebuttals to them. Having neglected the phenomenological dimension of consciousness, our proposal might be considered reductionist and incomplete. However, we believe our account provides a valuable tool for assessing the level of consciousness of an agent at least from a cognitive point of view.
Arrabales, R. Ledezma, A. and Sanchis, A. "A Multimodal Attention Mechanism for Autonomous Mobile Robotics". IX Workshop on Physical Agents 2008. Joaquín Lopez and Matías García (Eds.) Vigo. Spain. September 2008. Pages 121-128.
[PDF] [BibTeX] [Workshop Website] [Pics] Abstract
Whatever the mission of an autonomous mobile robot is, attention is a helpful cognitive capability when dealing with real world environments. In this paper we present a novel control architecture which enables an integrated and efficient filtering of multiple modality sensory information. The concept of context is introduced as the set of criteria that determines what sensory information is relevant to the current mission. The proposed attention mechanism uses these contexts as a mean to adaptively select the constrained cognitive focus of the robot within the vast multimodal sensory space available. This approach for artificial attention is tested in the domain of autonomous mapping.
Arrabales, R. Ledezma, A. and Sanchis, A. "Criteria for Consciousness in Artificial Intelligent Agents". ALAMAS+ALAg 2008 - Adaptive Learning Agents and Multi-Agent Systems - Proceedings of the ALAMAS+ALAg Workshop at AAMAS 2008. Estoril. Portugal. Pages 57-64. 2008.
[AAMAS: CORE ERA Conference Ranking: A]
[PDF] [BibTeX] [Workshop Website] Abstract
Accurately testing for consciousness is still an unsolved problem when applied to humans and other mammals. The inherent subjective nature of conscious experience makes it virtually unreachable to classic empirical approaches. Therefore, alternative strategies based on behavior analysis and neurobiological studies are being developed in order to determine the level of consciousness of biological organisms. However, these methods cannot be directly applied to artificial systems. In this paper we propose both a taxonomy and some functional criteria that can be used to assess the level of consciousness of an artificial intelligent agent. Furthermore, a list of measurable levels of artificial consciousness, ConsScale, is defined as a tool to determine the potential level of consciousness of an agent. Both the mapping of consciousness to AI and the role of consciousness in cognition are controversial and unsolved questions, in this paper we aim to approach these issues with the notions of IConsciousness and embodied intelligence.
Arrabales, R. and Sanchis de Miguel, A. "A Machine Consciousness Approach to Autonomous Mobile Robotics". In the 5th International Cognitive Robotics Workshop. AAAI-06. Boston, MA. July 2006.
[PDF] [BibTeX] [RefWorks] [ProCite/EndNote] [Workshop] Abstract
In this paper we argue that machine consciousness can be successfully modelled to be the base of a control system for an autonomous mobile robot. Such a bio-inspired system provides the robot with cognitive benefits the same way that consciousness does for humans and other higher mammals. The key functions of consciousness are identified and partially applied to an original computational model, which is implemented in a software simulated mobile robot. We use a simulator to prove our assumptions and gain insight about the benefits that conscious and affective functions add to the behaviour of the robot. A particular exploration problem is analyzed and experiments results are evaluated. We conclude that this cognitive approach involving consciousness and emotion functions cannot be ignored in the design of mobile robots, as it provides efficiency and robustness in autonomous tasks. Specifically, the proposed model has revealed efficient control behaviour when dealing with unexpected situations. Arrabales, R. Ledezma, A. and Sanchis, A. "CC-Bot: A Believable Video Game Bot based on Machine Consciousness". 5th International Conference on Cognitive Systems (CogSys 2012). Vienna. Austria. February 2012.
[PDF] [Conference Website] [Proceedings] Abstract
CC-Bot is a virtual agent designed to show human-like behavior during a deathmatch game in a First-Person shooter video game.
Arrabales, R. Ledezma, A. and Sanchis, A. "ConsScale: A Cognitive Scale Inspired on Consciousness". 4th International Conference on Cognitive Systems (CogSys 2010). Zürich. Switzerland. January 2010.
[PDF] [BibTeX] [Conference Website] [Proceedings] [Pics] Abstract
ConsScale is a framework for characterizing the cognitive power of a creature. ConsScale includes the definition of an ordered list of cognitive levels arranged across a developmental path. The arrangement of the levels is inspired on the ontogeny and phylogeny of consciousness in biological organisms. The basic assumption is that there exist different kinds of minds, and they can be characterized in terms of ConsScale criteria. Using ConsScale, characterization and assessment of consciousness can be performed using three related tools: the conceptual levels of consciousness, the CQS (ConsScale Quantitative Score), and the ConsScale radar graph representation. In order to assess the level of artificial consciousness of an agent using ConsScale, its architectural components have to be identified and its cognitive skills tested. Using this information as input, the scale can be used to obtain both a qualitative and a quantitative measure of consciousness.
Arrabales, R. Ledezma, A. and Sanchis, A. "CRANIUM-CERA Cognitive Architecture". 1st International Semminar on New Issues of Artificial Intelligence. Universidad Carlos III de Madrid. February 2008.
[Poster] [Semminar Website] Arrabales, R. Muñoz, J. "¿Máquina o Humano?". Revista Investigación y Ciencia (Spanish edition of Scientific American). No. 416. Sección Panorama. Mayo 2011.
[Website] Abstract La consciencia artificial constituye un campo de investigación multidisciplinar cuyo objetivo consiste en comprender los mecanismos que originan la consciencia en el cerebro humano y replicarlos en dispositivos artificiales. Una de sus numerosas aplicaciones se orienta hacia el desarrollo de agentes artificiales, como robots autónomos o personajes de videojuegos, cuyo comportamiento emule al de un ser humano.
Para decidir si la conducta de una máquina resulta tan inteligente como la de un humano, el matemático Alan Turing describió hace más de 60 años la prueba que hoy se conoce como test de Turing. En su versión original, consiste en pedir a una persona, que actúa a modo de «juez», que converse por escrito y mediante un terminal con dos interlocutores sin identificar: una máquina y un ser humano. Si, después de charlar con ambos durante un tiempo razonable, el juez no logra adivinar quién es quién, la máquina puede ser considerada tan inteligente como una persona. Hasta la fecha, ninguna máquina ha superado el test de Turing.
Arrabales, R. Muñoz, J. "Conscious-like Bots wins the 2K BotPrize". ERCIM News, No. 84. Special theme: Intelligent and cognitive systems. January 2011.
[PDF] [Website] Abstract CCBot-2, a software agent based on the CERA-CRANIUM cognitive architecture, was the winner of this year’s edition of the BotPrize. This contest is an adaptation of the Turing test for a first person shooter video game. Although CCBot-2 could not pass the Turing test, she narrowed the gap with human players, being considered the most human bot, and achieving a humanness ratio of almost 32% (while the “less human” human player scored around 35%).
Arrabales, R. "El mundo interior de PECA". V Certamen "Teresa Pinillos" de divulgación científica y humanística. Noviembre 2010.
[PDF] [Website] Abstract Como buen científico no dejo de hacerme preguntas sobre la realidad que nos rodea. Las preguntas que a menudo me parecen más interesantes están relacionadas con las fronteras de la propia ciencia. ¿Puede proporcionarnos la ciencia la verdad acerca de todas las cosas? ¿Dónde está el límite? ¿Es aplicable el método científico a todos los fenómenos conocidos?.
Arrabales, R. and Muñoz, J. "The Awakening of Conscious Bots: Inside the Mind of the 2K BotPrize 2010 Winner". AIGameDev. October 2010.
[PDF] [Website] Abstract This in-depth article about CC-Bot2, the winning entry of this year's 2K Bot Prize at CIG '10, was written by Raúl Arrabales and Jorge Muñoz, also known as the Conscious-Robots team.
Arrabales, R. "El Ataque de los Cylon II". Blog de Sistemas Inteligentes de Madri+d. Agosto 2009.
[PDF] [Website] Abstract Seguimos la serie de posts titulada “El Ataque de los Cylon“. En este artículo contamos con la colaboración de Raúl Arrabales, uno de los mayores expertos nacionales en el tema de la Consciencia Artificial, y gestor del magnífico portal sobre consciencia y robótica ConsciousRobots, que a lo largo del post discute la evolución de los robots conscientes, y analiza las posibilidades a medio y largo plazo de los mismos, estudiando el momento en el que podremos contar con robots conscientes en una forma similar a la humana, el equivalente a los Cylon de Battlestar Galactica. A continuación tenéis el excelente post de Raúl, al que agradecemos el tiempo y la dedicación a contribuir a esta serie de posts, que esperamos sean de vuestro interés.
|
