Menu
Home
People
Places
Arts
History
Plants & Animals
Science
Life & Culture
Technology
Reference.org
OpenCog
open-in-new
<h2 id="origin">Origin</h2> <p>OpenCog was originally based on the release in 2008 of the source code of the proprietary "Novamente Cognition Engine" (NCE) of Novamente LLC. The original NCE code is discussed in the PLN book (ref below). Ongoing development of OpenCog is supported by Artificial General Intelligence Research Institute (AGIRI), the <a href="/facts/Google_Summer_of_Code/M8rDIn6W">Google Summer of Code</a> project, <a href="/facts/Hanson_Robotics/lwrbgSnS">Hanson Robotics</a>, SingularityNET and others. </p> <h2 id="components">Components</h2> <p>OpenCog consists of: </p> <ul><li>A <a href="/facts/Graph_database/WJfPlV3z">graph database</a>, dubbed the <a href="https://wiki.opencog.org/w/Atomspace">AtomSpace</a>, that holds "atoms" (that is, <a href="/facts/Term_(logic)/M0nXp7dH">terms</a>, <a href="/facts/Atomic_formula/s9WMnbqT">atomic formulas</a>, <a href="/facts/Sentence_(mathematical_logic)/lnsLzvTx">sentences</a> and <a href="/facts/Relational_model/VFf4dDIU">relationships</a>) together with their "values" (<a href="/facts/Valuation_(logic)/FWhmAhy9">valuations</a> or <a href="/facts/Interpretation_(logic)/goPmjjQG">interpretations</a>, which can be thought of as per-atom <a href="/facts/Key-value_database/Yee8EIGE">key-value databases</a>). An example of a value would be a <a href="/facts/Truth_value/SY4e2w8l">truth value</a>. Atoms are globally unique, immutable and are indexed (searchable); values are fleeting and changeable.</li></ul> <ul><li>A collection of pre-defined atoms, termed <a href="https://wiki.opencog.org/w/Atomese">Atomese</a>, used for generic <a href="/facts/Knowledge_representation/9lnzyILD">knowledge representation</a>, such as <a href="/facts/Conceptual_graph/8cHXdCae">conceptual graphs</a> and <a href="/facts/Semantic_network/mt7FlP53">semantic networks</a>, as well as to represent and store the rules (in the sense of <a href="/facts/Term_rewriting/9PHlnnnJ">term rewriting</a>) needed to manipulate such graphs.</li></ul> <ul><li>A collection of pre-defined atoms that encode a <a href="/facts/Type_theory/xj0xzVqI">type subsystem</a>, including <a href="/facts/Type_constructor/4zRCrxUW">type constructors</a> and <a href="/facts/Function_type/CmbusBzk">function types</a>. These are used to specify the types of variables, terms and expressions, and are used to specify the structure of generic graphs containing variables.</li></ul> <ul><li>A collection of pre-defined atoms that encode both <a href="/facts/Functional_programming_language/cdxqPEIM">functional</a> and <a href="/facts/Imperative_programming_language/6b5dUnE1">imperative</a> programming styles. These include the <a href="/facts/Lambda_abstraction/DaD3RuQ4">lambda abstraction</a> for binding <a href="/facts/Free_variable/ndWnXXYs">free variables</a> into <a href="/facts/Bound_variable/ndWnXXYs">bound variables</a>, as well as for performing <a href="/facts/Beta_reduction/DaD3RuQ4">beta reduction</a>.</li></ul> <ul><li>A collection of pre-defined atoms that encode a <a href="/facts/Satisfiability_modulo_theories/10mBFLFB">satisfiability modulo theories</a> solver, built in as a part of a generic graph query engine, for performing graph and hypergraph pattern matching (<a href="/facts/Graph_isomorphism/SBjVIhwW">isomorphic subgraph discovery</a>). This generalizes the idea of a structured query language (<a href="/facts/SQL/UjQLJcYm">SQL</a>) to the domain of generic graphical queries; it is an extended form of a graph <a href="/facts/Query_language/Sa6K4xXM">query language</a>.</li></ul> <ul><li>A generic <a href="/facts/Rule_engine/3mem1gI0">rule engine</a>, including a <a href="/facts/Forward_chaining/FJjYIf6S">forward chainer</a> and a <a href="/facts/Backward_chaining/St5dtGYq">backward chainer</a>, that is able to chain together rules. The rules are exactly the graph queries of the graph query subsystem, and so the rule engine vaguely resembles a <a href="/facts/Query_plan/VDsyMQFE">query planner</a>. It is designed so as to allow different kinds of <a href="/facts/Inference_engine/oLlNX14Q">inference engines</a> and <a href="/facts/Reasoning_system/V6c6TToD">reasoning systems</a> to be implemented, such as <a href="/facts/Bayesian_inference/tEpK3zLx">Bayesian inference</a> or <a href="/facts/Fuzzy_logic/1PvN4Att">fuzzy logic</a>, or practical tasks, such as <a href="/facts/Constraint_solver/rgkCK5bD">constraint solvers</a> or <a href="/facts/Planning/ckzCp5tv">motion planners</a>.</li></ul> <ul><li>An attention allocation subsystem based on economic theory, termed ECAN.<a class="footnote-ref" id="fnref:3" href="#fn:3"><sup>3</sup></a> This subsystem is used to control the combinatorial explosion of search possibilities that are met during inference and chaining.</li></ul> <ul><li>An implementation of a probabilistic <a href="/facts/Reasoning_system/V6c6TToD">reasoning engine</a> based on <a href="/facts/Probabilistic_logic_network/3aF7FFkK">probabilistic logic networks</a> (PLN). The current implementation uses the rule engine to chain together specific rules of <a href="/facts/Logical_inference/RBbe95bP">logical inference</a> (such as <a href="/facts/Modus_ponens/Wq4NcmrG">modus ponens</a>), together with some very specific mathematical formulas assigning a probability and a confidence to each deduction. This subsystem can be thought of as a certain kind of <a href="/facts/Proof_assistant/hjFGxLPn">proof assistant</a> that works with a modified form of Bayesian inference.</li></ul> <ul><li>A probabilistic <a href="/facts/Genetic_programming/kGa2jr0w">genetic program evolver</a> called Meta-Optimizing Semantic Evolutionary Search, or MOSES.<a class="footnote-ref" id="fnref:4" href="#fn:4"><sup>4</sup></a> This is used to discover collections of short Atomese programs that accomplish tasks; these can be thought of as performing a kind of <a href="/facts/Decision_tree_learning/LbPMAwdI">decision tree learning</a>, resulting in a kind of <a href="/facts/Random_forest/Ff1ppJEQ">decision forest</a>, or rather, a generalization thereof.</li></ul> <ul><li>A natural language input system consisting of <a href="/facts/Link_Grammar/2cSELk8L">Link Grammar</a>, and partly inspired by both <a href="/facts/Meaning-Text_Theory/ptm2fvN3">Meaning-Text Theory</a> as well as Dick Hudson's <a href="/facts/Word_grammar/wKfP9Nrk">Word Grammar</a>, which encodes semantic and syntactic relations in Atomese.</li></ul> <ul><li>A natural language generation system.<a class="footnote-ref" id="fnref:5" href="#fn:5"><sup>5</sup></a></li></ul> <ul><li>An implementation of <a href="/facts/Psi-Theory/VydhyDUs">Psi-Theory</a> for handling <a href="/facts/Emotion/kvAr22wa">emotional</a> states, drives and urges, dubbed OpenPsi.<a class="footnote-ref" id="fnref:6" href="#fn:6"><sup>6</sup></a></li></ul> <ul><li>Interfaces to <a href="/facts/Hanson_Robotics/lwrbgSnS">Hanson Robotics</a> robots, including emotion modelling<a class="footnote-ref" id="fnref:7" href="#fn:7"><sup>7</sup></a> via OpenPsi. This includes the Loving AI project, used to demonstrate meditation techniques.</li></ul> <h2 id="organization-and-funding">Organization and funding</h2> <p>In 2008, the <a href="/facts/Machine_Intelligence_Research_Institute/LfTdXVp9">Machine Intelligence Research Institute</a> (MIRI), formerly called Singularity Institute for Artificial Intelligence (SIAI), sponsored several researchers and engineers. Many contributions from the open source community have been made since OpenCog's involvement in the <a href="/facts/Google_Summer_of_Code/M8rDIn6W">Google Summer of Code</a> in 2008 and 2009. Currently MIRI no longer supports OpenCog.<a class="footnote-ref" id="fnref:8" href="#fn:8"><sup>8</sup></a> OpenCog has received funding and support from several sources, including the Hong Kong government, <a href="/facts/Hong_Kong_Polytechnic_University/CHjvYpPE">Hong Kong Polytechnic University</a>, the <a href="/facts/Jeffrey_Epstein_VI_Foundation/1wbRtCu4">Jeffrey Epstein VI Foundation</a><a class="footnote-ref" id="fnref:9" href="#fn:9"><sup>9</sup></a> and <a href="/facts/Hanson_Robotics/lwrbgSnS">Hanson Robotics</a>. The OpenCog project is currently affiliated with SingularityNET and Hanson Robotics. </p> <h2 id="applications">Applications</h2> <p>Similar to other <a href="/facts/Cognitive_architecture/6EgYFrx5">cognitive architectures</a>, the main purpose is to create virtual humans, which are three dimensional <a href="/facts/Avatar_(computing)/AVTBXnrl">avatar</a> characters. The goal is to mimic behaviors like emotions, gestures and learning. For example, the emotion module in the software was only programmed because humans have emotions. Artificial General Intelligence can be realized if it simulates intelligence of humans.<a class="footnote-ref" id="fnref:10" href="#fn:10"><sup>10</sup></a> </p><p>The self-description of the OpenCog project provides additional possible applications which are going into the direction of natural language processing and the simulation of a dog.<a class="footnote-ref" id="fnref:11" href="#fn:11"><sup>11</sup></a> </p> <h2 id="see-also">See also</h2> <ul><li><a href="/facts/Soar_(cognitive_architecture)/jV7B7EzH">Soar</a></li> <li><a href="/facts/Cyc/JyTurpR6">Cyc</a></li> <li><a href="/facts/List_of_free_and_open-source_software_packages/l3x4PXwn">List of free and open source software packages</a></li> <li><a href="/facts/OpenAI/V7WVK1t4">OpenAI</a></li></ul> <h2 id="sources">Sources</h2> <ul><li>Hart, D; B Goertzel (2008). <a href="http://www.agiri.org/OpenCog_AGI-08.pdf"><i>OpenCog: A Software Framework for Integrative Artificial General Intelligence</i></a> (PDF). Proceedings of the First AGI Conference. <a href="https://books.google.com/books?id=a_ZR81Z25z0C">Gbooks</a></li></ul> <ul><li>Goertzel, B., Iklé, M., Goertzel, I.F., Heljakka, A. <i><a href="https://www.springer.com/computer/artificial/book/978-0-387-76871-7">Probabilistic Logic Networks, A Comprehensive Framework for Uncertain Inference</a></i>, Springer, 2009, VIII, 336 p., Hardcover <a href="/facts/ISBN_(identifier)/15AdSPa9">ISBN</a> 978-0-387-76871-7</li></ul> <h2 id="external-links">External links</h2> <ul><li><a href="http://opencog.org">Official website</a></li> <li><a href="http://wiki.opencog.org/w/The_Open_Cognition_Project">OpenCog Wiki</a></li> <li><a href="https://www.youtube.com/watch?v=x18yaOXBSQA">AGI 2011: OpenCog - GoogleTechTalks</a> on <a href="/facts/YouTube_video_(identifier)/db276jst">YouTube</a></li> <li><a href="https://www.youtube.com/watch?v=Rgjw8O3vLBs">AGI 2011: Architectures Part I - GoogleTechTalks</a> on <a href="/facts/YouTube_video_(identifier)/db276jst">YouTube</a></li> <li><a href="https://www.youtube.com/watch?v=5hsvCib83ME">Artificial General Intelligence: Now Is the Time - 2007 GoogleTechTalks</a> on <a href="/facts/YouTube_video_(identifier)/db276jst">YouTube</a></li> <li><a href="http://wiki.opencog.org/w/CogPrime_Overview">CogPrime: An Integrative Architecture for Embodied Artificial General Intelligence</a></li> <li><a href="https://www.youtube.com/watch?v=qfwMV18is6M">OpenCog: An Open Source Software Framework & A Design & Vision for Advanced AGI. Video</a> on <a href="/facts/YouTube_video_(identifier)/db276jst">YouTube</a> Given at Monash University Australia, Sept 2011. Adam Ford</li> <li><a href="https://www.youtube.com/watch?v=mdLbg-anczk">Video introduction to OpenCog by Ben Goertzel Video</a> on <a href="/facts/YouTube_video_(identifier)/db276jst">YouTube</a>. Ben speaks on OpenCog in Tai Po, Hong Kong, Dec 2011. Adam Ford</li> <li><a href="https://www.youtube.com/watch?v=4xogjUqEQvY">Ben Goertzel - the future of AGI - Open Cog development in Asia Video</a> on <a href="/facts/YouTube_video_(identifier)/db276jst">YouTube</a> Adam Ford</li></ul> <h2 id="references">References</h2> <ol> <li id="fn:1"><p>"OpenCog: Open-Source Artificial General Intelligence for Virtual Worlds | CyberTech News". 2009-03-06. Archived from the original on 2009-03-06. Retrieved 2016-10-01.{{cite web}}: CS1 maint: bot: original URL status unknown (link) <a href="https://web.archive.org/web/20090306053354/http://www.cybertechnews.org/?p=915" target="_blank">https://web.archive.org/web/20090306053354/http://www.cybertechnews.org/?p=915</a> <a href="#fnref:1" class="footnote-back-ref">↩</a></p></li> <li id="fn:2"><p>Rogers, Stewart (2017-12-07). "SingularityNET talks collaborative AI as its token sale hits 400% oversubscription". venturebeat.com. VentureBeat. Retrieved 2018-03-13. <a href="https://venturebeat.com/2017/12/07/singularitynet-talks-collaborative-ai-as-its-token-sale-hits-400-oversubscription/" target="_blank">https://venturebeat.com/2017/12/07/singularitynet-talks-collaborative-ai-as-its-token-sale-hits-400-oversubscription/</a> <a href="#fnref:2" class="footnote-back-ref">↩</a></p></li> <li id="fn:3"><p>"Economic Attention Allocation". <a href="https://wiki.opencog.org/w/ECAN" target="_blank">https://wiki.opencog.org/w/ECAN</a> <a href="#fnref:3" class="footnote-back-ref">↩</a></p></li> <li id="fn:4"><p>"MOSES". <a href="https://wiki.opencog.org/w/MOSES" target="_blank">https://wiki.opencog.org/w/MOSES</a> <a href="#fnref:4" class="footnote-back-ref">↩</a></p></li> <li id="fn:5"><p>"Natural Language Generation". <a href="https://wiki.opencog.org/w/Natural_language_generation" target="_blank">https://wiki.opencog.org/w/Natural_language_generation</a> <a href="#fnref:5" class="footnote-back-ref">↩</a></p></li> <li id="fn:6"><p>"OpenPsi". <a href="https://wiki.opencog.org/w/OpenPsi" target="_blank">https://wiki.opencog.org/w/OpenPsi</a> <a href="#fnref:6" class="footnote-back-ref">↩</a></p></li> <li id="fn:7"><p>"Emotion modeling - Hanson Robotics Wiki". Archived from the original on 2018-03-19. Retrieved 2015-04-24. <a href="https://web.archive.org/web/20180319173932/http://wiki.hansonrobotics.com/w/Emotion_modeling" target="_blank">https://web.archive.org/web/20180319173932/http://wiki.hansonrobotics.com/w/Emotion_modeling</a> <a href="#fnref:7" class="footnote-back-ref">↩</a></p></li> <li id="fn:8"><p>Ben Goertzel (2010-10-29). "The Singularity Institute's Scary Idea (and Why I Don't Buy It)". The Multiverse According to Ben. Retrieved 2011-06-24. <a href="/wiki/Ben_Goertzel" target="_blank">/wiki/Ben_Goertzel</a> <a href="#fnref:8" class="footnote-back-ref">↩</a></p></li> <li id="fn:9"><p>Baram, Marcus (Jul 11, 2019). "Even after his arrest, scientists were more than happy to take money from Jeffrey Epstein". Fast Company. <a href="https://www.fastcompany.com/90375335/jeffrey-epsteins-money-was-accepted-by-scientists-even-after-arrest" target="_blank">https://www.fastcompany.com/90375335/jeffrey-epsteins-money-was-accepted-by-scientists-even-after-arrest</a> <a href="#fnref:9" class="footnote-back-ref">↩</a></p></li> <li id="fn:10"><p>David Burden; Maggi Savin-Baden (24 January 2019). Virtual Humans: Today and Tomorrow. CRC Press. ISBN 978-1-351-36526-0. Retrieved 25 August 2020. <a href="978-1-351-36526-0" target="_blank">978-1-351-36526-0</a> <a href="#fnref:10" class="footnote-back-ref">↩</a></p></li> <li id="fn:11"><p>Ben Goertzel; Cassio Pennachin; Nil Geisweiller (8 July 2014). Engineering General Intelligence, Part 1: A Path to Advanced AGI via Embodied Learning and Cognitive Synergy. Springer. pp. 23–. ISBN 978-94-6239-027-0. <a href="978-94-6239-027-0" target="_blank">978-94-6239-027-0</a> <a href="#fnref:11" class="footnote-back-ref">↩</a></p></li> </ol>