LIDA, (Learning IDA), cognitive architecture, developed under Stanley P. Franklin, Cognitive Computing Research Group (CCRG) at the University of Memphis, Memphis, TN, USA.
Email: Этот адрес электронной почты защищён от спам-ботов. У вас должен быть включен JavaScript для просмотра. | Stanley P. Franklin - Home Page
LIDA Home Page | LIDA architecture Cognitive Computing Research Group
People | Projects | Tutorial | Framework | LIDA wiki
Когнитивная архитектура LIDA разрабатывается под руководством Стэнли Франклина (Stan Franklin) исследовательской группой по проекту в Университете Мемфиса (University of Memphis), США.
Общие сведения
The LIDA (Learning Intelligent Distribution Agent) cognitive architecture
The LIDA (Learning Intelligent Distribution Agent) cognitive architecture is an integrated artificial cognitive system that attempts to model a broad spectrum of cognition in biological systems, from low-level perception/action to high-level reasoning. Developed primarily by Stan Franklin and colleagues at the University of Memphis, the LIDA architecture is empirically grounded in cognitive science and cognitive neuroscience.
In addition to providing hypotheses to guide further research, the architecture can support control structures for software agents and robots.
Providing plausible explanations for many cognitive processes, the LIDA conceptual model is also intended as a tool with which to think about how minds work.
Two hypotheses underlie the LIDA architecture and its corresponding conceptual model:
1) Much of human cognition functions by means of frequently iterated (~10 Hz) interactions, called cognitive cycles, between conscious contents, the various memory systems and action selection.
2) These cognitive cycles, serve as the “atoms” of cognition of which higher-level cognitive processes are composed.
Computational mechanisms
The LIDA architecture employs several modules that are designed using computational mechanisms drawn from the “new AI.” These include variants of the Copycat Architecture, Sparse Distributed Memory, the Schema Mechanism, the Behavior Net, and the Subsumption Architecture.
Psychological and neurobiological underpinnings
As a comprehensive, conceptual and computational cognitive architecture the LIDA architecture is intended to model a large portion of human cognition. Comprising a broad array of cognitive modules and processes, the LIDA architecture attempts to implement and flesh out a number of psychological and neuropsychological theories including Global Workspace Theory, Situated Cognition, perceptual symbol systems, Working Memory, memory by affordances, long-term working memory, and the H-CogAff architecture.
LIDA’s cognitive cycle
The LIDA cognitive cycle can be subdivided into three phases:
the understanding phase,
the attention (consciousness) phase,
and the action selection and learning phase.
Beginning the understanding phase, incoming stimuli activate low-level feature detectors in Sensory Memory. The output engages Perceptual Associative Memory where higher-level feature detectors feed in to more abstract entities such as objects, categories, actions, events, etc. The resulting percept moves to the Workspace where it cues both Transient Episodic Memory and Declarative Memory producing local associations. These local associations are combined with the percept to generate a current situational model; the agent’s understanding of what is going on right now.
The attention phase begins with the forming of coalitions of the most salient portions of the current situational model, which then compete for attention, that is a place in the current conscious contents. These conscious contents are then broadcast globally, initiating the learning and action selection phase. New entities and associations, and the reinforcement of old ones, occur as the conscious broadcast reaches the various forms of memory, perceptual, episodic and procedural.
In parallel with all this learning, and using the conscious contents, possible action schemes are instantiated from Procedural Memory and sent to Action Selection, where they compete to be the behavior selected for this cognitive cycle. The selected behavior triggers Sensory-Motor Memory to produce a suitable algorithm for its execution, which completes the cognitive cycle.
Figure 1. The LIDA Cognitive Cycle Diagram
A more recent LIDA cogntive cycle diagram can be downloaded as a zip file.
LIDA Software Framework
Work on a Software Framework for LIDA agents began in January 2009. As a software framework it promotes code reuse providing new users with a usable system instead of having to start from scratch. It allows developers to apply design principles, patterns and best practices easily. As a consequence the LIDA software framework is highly customizable. Essentially the framework implements the common, well-understood parts of the architecture leaving room for current and future developers to concentrate on domain (and module) specific problems. Cognitive systems can be very complex but they typically have a defined architecture with many domain independent parts.
The LIDA framework is intended to create a generic and configurable version of the domain independent modules and processes of LIDA implemented as a software framework in Java. It should be easily customizable for different domains (environments). It should permit changes to the implementation of each module and allow for XML definitions of the data structures, processes, parameters, and modules of the system.
Since LIDA is a psychologically-realistic cognitive system, the framework should embrace parallelism and employ multithreading. Also, it should eschew an "information processing" style cognition in favor of many interacting process operating asynchronously.
Finally, as a complex system, it should provide users with tools to make it more mangeable: GUI display of the inner workings of the system, logging of important events, and utilities to parse XML and Properties files for easy customization.
LIDA Software Framework
LIDA Framework Software Non-Exclusive, Non-Commercial Use License (pdf)
May, 2012 - An update to the LIDA software framework, version 1.2 beta, is now available for download!
Check it out here.
Work on a Software Framework for LIDA agents began in January 2009. As a software framework it promotes code reuse providing new users with a usable system instead of having to start from scratch. It allows developers to apply design principles, patterns and best practices easily. As a consequence the LIDA software framework is highly customizable. Essentially the framework implements the common, well-understood parts of the architecture leaving room for current and future developers to concentrate on domain (and module) specific problems. Cognitive systems can be very complex but they typically have a defined architecture with many domain independent parts.
The LIDA framework is intended to create a generic and configurable version of the domain independent modules and processes of LIDA implemented as a software framework in Java. It should be easily customizable for different domains (environments). It should permit changes to the implementation of each module and allow for XML definitions of the data structures, processes, parameters, and modules of the system.
Since LIDA is a psychologically-realistic cognitive system, the framework should embrace parallelism and employ multithreading. Also, it should eschew an "information processing" style cognition in favor of many interacting process operating asynchronously.
Finally, as a complex system, it should provide users with tools to make it more mangeable: GUI display of the inner workings of the system, logging of important events, and utilities to parse XML and Properties files for easy customization.
Internet Groups and Mailing Lists:
ccrg-memphis Cognitive Computing Research Group - CCRG. This is a mailing list for the Cognitve Computing Research Group at the University of Memphis. Members are welcome to ask questions and start topics about the LIDA Cognitive Model, the LIDA Software Framework, or related topics
To post a message to all the list members, send email to Этот адрес электронной почты защищён от спам-ботов. У вас должен быть включен JavaScript для просмотра.
Publications
Snaider, J., McCall, R., & Franklin, S. (2011). The LIDA Framework as a General Tool for AGI. The Fourth Conference on Artificial General Intelligence. (View: PDF)
Franklin, S. (2011). Global Workspace Theory, Shanahan, and LIDA. International Journal of Machine Consciousness, 3(2), 327-337. doi: 10.1142/S1793843011000728 (View: PDF)
McCall, R., Franklin, S., Friedlander, D. (2010). Grounded Event-Based and Modal Representations for Objects, Relations, Beliefs, Etc. Paper presented at the FLAIRS-23, Daytona Beach, FL.
(View: PDF)
Franklin, S. (2014). History, motivations and core themes of AI. In K. Frankish & W. Ramsey (Eds.), The Cambridge Handbook of Artificial Intelligence (pp. 15-33). Cambridge: Cambridge University Press. (View: PDF)
Agrawal, P., & Franklin, S. (2014 to appear). Multi-Layer Cortical Learning Algorithms. Paper presented at the IEEE Symposium Series on Computational Intelligence (SSCI). (View Draft: PDF)
Franklin, S., Madl, T., D'Mello, S., & Snaider, J. (2014). LIDA: A Systems-level Architecture for Cognition, Emotion, and Learning. IEEE Transactions on Autonomous Mental Development, 6(1), 19-41.
doi: 10.1109/TAMD.2013.2277589 (View: PDF)
Franklin, S., Strain, S., McCall, R., & Baars, B. (2013). Conceptual Commitments of the LIDA Model of Cognition. Journal of Artificial General Intelligence, 4(2), 1-22, DOI: 10.2478/jagi-2013-0002 (View: PDF)
Madl, T., Franklin, S., Chen, K., & Trappl, R. (2013). Spatial Working Memory in the LIDA Cognitive Architecture. In R. West & T. Stewart (Eds.), Proceedings of the 12th International Conference on Cognitive Modelling (pp. 384-390). Ottawa, Canada: Carleton University. (View: PDF)
Faghihi, U., McCall, R., & Franklin, S. (2012). A Computational Model of Attentional Learning in a Cognitive Agent. Biologically Inspired Cognitive Architectures, 2, 25-36. (View: PDF)
McCall, R. & Franklin, S. (2012). Meta Learning, Change of Internal Workings, and LIDA: A commentary on Thórisson and Helgasson's "Cognitive architectures and autonomy: A Comparative Review" [Peer commentary by R. McCall & S. Franklin]. Journal of Artificial General Intelligence, 3(2), 42-44, DOI: 10.2478/v10229-011-0016-2 (View: PDF)
Madl, T., Baars, B. J., & Franklin, S. (2011). The Timing of the Cognitive Cycle. PLoS ONE, 6(4), e14803. (View: Online, PDF)
Digital Edition of A Cognitive Theory of Consciousness by Bernard Baars
Bogner, M., J. Maletic, & S. Franklin. ConAg: a reusable framework for developing "conscious" software agents.
(View: Abstract or PDF)
McCauley, L., S. Franklin , and M. Bogner. 2000. An Emotion-Based "Conscious" Software Agent Architecture. In Affective Interactions, Lecture Notes on Artificial Intelligence ed., vol. 1814, ed. A. Paiva. Berlin : Springer.
Franklin, S., A. Kelemen, and L. McCauley. 1998. IDA: A Cognitive Agent Architecture. In IEEE Conf on Systems, Man and Cybernetics. IEEE Press.
Franklin, Stan, Song, Hongjun, and Negatu, Aregahegn, SUMPY: A Fuzzy Software Agent ed. F. C. Harris, Jr., Intelligent Systems: Proceedings of the ISCA 5th International Conference, (Reno Nevada, June 1996) Raleigh NC : International Society for Computers and Their Applications - ISCA, 124-129.
Franklin, Stan, Autonomous Agents as Embodied AI Cybernetics and Systems, special issue on Epistomological Issues in Embodied AI, 28:6 1997 499-520.
David Friedlander, Stan Franklin
Baars, B. J. 1988. A Cognitive Theory of Consciousness. Cambridge University Press. (PDF)
Bernard J. Baars and Nicole M. Gage. Cognition, Brain, and Consciousness, the 2nd Edition: Introduction to Cognitive Neuroscience (html)
