brain-inspired cognitive architectures włodzisław duch department of informatics, nicolaus...
out of 86
Post on 24-Dec-2015
Embed Size (px)
- Slide 1
- Brain-Inspired Cognitive Architectures Wodzisaw Duch Department of Informatics, Nicolaus Copernicus University, Toru, Poland Google: W. Duch IRML Summer School, NYU Abu Dhabi, 7/2012
- Slide 2
- Slide 3
- Copernicus Nicolaus Copernicus: born in Torun in 1472
- Slide 4
- Singapore Nanyang Technological University
- Slide 5
- DI NCU Projects: CI Google W. Duch => List of projects, talks, papers Computational intelligence (CI), main themes: Foundations of computational intelligence: transformation based learning, k-separability, learning hard boolean problems. Meta-learning, or learning how to learn. Understanding of data: prototype-based rules, visualization. Novel learning: projection pursuit networks, QPC (Quality of Projected Clusters), search-based neural training, transfer learning or learning from others (ULM), aRPM, SFM... Similarity based framework for metalearning, heterogeneous systems, new transfer functions for neural networks. Feature selection, extraction, creation, transfer learning.
- Slide 6
- Slide 7
- DI NCU Projects:NCI Neurocognitive Informatics projects. Computational creativity, insight, intuition, consciousness. Neurocognitive approach to language, word games. Medical information retrieval, analysis, visualization. Global analysis of EEG, visualization of high-D trajectories. Brain stem models and consciousness in artificial systems. Autism, comprehensive theory. Imagery agnosia, especially imagery amusia. Infants: observation, guided development. A test-bed for integration of different Humanized Interface Technologies (HIT), with Singapore C 2 I Center. Free will, neural determinism and social consequences.
- Slide 8
- A roadmap to human level intelligence workshop organized by: Wodzisaw Duch (Google: W. Duch) Department of Informatics, Nicolaus Copernicus University, Torun, Poland & School of Computer Engineering, Nanyang Technological Uni, Singapore Nikola Kasabov (http://www.kedri.info) KEDRI, Auckland, New Zealand James Anderson, Paul Allopenna, Robert Hecht-Nielsen, Andrew Coward, Alexei Samsonovich, Giorgio Ascoli, Kenneth De Jong, Ben Goertzel WCCI2006, Vancouver,, British Columbia, Canada, July 17, 2006
- Slide 9
- Steps Toward an AGI Roadmap Wodek Duch ( Google: W. Duch) Artificial General Intelligence (AGI, 2007 Memphis): architectures that can solve many problems and transfer knowledge between the tasks. Roadmaps: A Ten Year Roadmap to Machines with Common Sense (Push Singh, Marvin Minsky, 2002) A Ten Year Roadmap to Machines with Common Sense (Push Singh, Marvin Minsky, 2002) Euron (EU Robotics) Research Roadmap (2004) Euron (EU Robotics) Research Roadmap (2004) Neuro-IT Roadmap (EU, A. Knoll, M de Kamps, 2006) Neuro-IT Roadmap (EU, A. Knoll, M de Kamps, 2006) Challenges: Word games of increasing complexity: 20Q is the simplest, only object description. 20Q is the simplest, only object description. Yes/No game to understand situation. Yes/No game to understand situation. Logical entailment competitions. Logical entailment competitions. Conference series, journal, movement. Towards Human-like Intelligence, IEEE Computational Intelligence Society Task Force (J. Mandziuk & W. Duch), 2013 IEEE Symposium IEE
- Slide 10
- 2012 Initiatives IEEE Computational Intelligence Society Task Force (J. Mandziuk & W. Duch), Towards Human-like Intelligence. Brain-Mind InstituteBrain-Mind Institute School (25.06-3.08.2012), International Conference on Brain- Mind (ICBM) and Brain-Mind Magazine (Juyang Weng, Michigan SU). AGI: conference, Journal of Artificial General Intelligence comments on Cognitive Architectures and Autonomy: A Comparative Review (special issue, eds. Tan A-H, Franklin S, Duch W). BICA: Annual International Conf on Biologically Inspired Cognitive Architectures, 3rd Annual Meeting of the BICA Society, Palermo, Italy, 31.10- 3.11.2012
- Slide 11
- EU FP7 FET Flagships I 6 candidates for FET (Future Emerging Technologies) Flagships Projects, each a 10 year 1 billion EUR programs, in preparatory phase. Flagships Projects 2 winning projects will be announced in July 2012. FuturICT: Knowledge Accelerator and Crisis-Relief System: Unleashing the Power of Information for a Sustainable Future. What if global scale computing facilities were available that could analyse most of the data available in the world? ITFoM: The IT Future of Medicine, proposes a data-driven, individualised medicine of the future, based on the molecular/physiological/anatomical data from individual patients. ITFoM shall make general models of human pathways, tissues, diseases and ultimately of the human as a whole. Graphene-CA: Graphene Science and technology for ICT and beyond, electronics, spintronics, photonics, plasmonics
- Slide 12
- FET Flagships II HBP-PS: The Human Brain Project, understanding the way the human brain works could be key to enabling a whole range of brain related or inspired developments in Information and Communication Technologies, as well as having transformational implications for neuroscience and medicine. CA-RoboCom: Robot Companions for Citizens are soft skinned and sentient machines designed to deliver assistance to people. This assistance is defined in the broadest possible sense and covers all sorts of different settings. Based on multidisciplinary science and engineering, CA-RoboCom aims to develop a radically new approach towards machines and how these are deployed in society. Guardian Angels: Guardian Angels for a Smarter Planet. Providing Information and Communication Technologies to assist people in all sorts of complex situations is the long term goal of the Flagship Initiative, Guardian Angels.
- Slide 13
- FuturICT The FuturICT flagship proposal intends to unify hundreds of the best scientists in Europe to explore social life on earth and everything it relates to. FuturICT flagship proposal will produce historic breakthroughs and provide powerful new ways to manage challenges that make the modern world so difficult to predict, including the financial crisis. A bit like Cybersyn project in Chile (1970-73) in real-time cybernetics control of economy.Cybersyn
- Slide 14
- Failures of AI Many ambitious general AI projects failed, for example: A. Newell, H. Simon, General Problem Solver (1957). Eduardo Caianiello (1961) mnemonic equations explain everything. 5 th generation computer project 1982-1994. AI has failed in many areas: problem solving, reasoning flexible control of behavior perception, computer vision language... Why? Too naive? Not focused on applications? Not addressing real challenges?
- Slide 15
- Ambitious approaches CYC, started by Douglas Lenat in 1984, commercial since 1995. Developed by CyCorp, with 2.5 millions of assertions linking over 150.000 concepts and using thousands of micro-theories (2004). Cyc-NL is still a potential application, knowledge representation in frames is quite complicated and thus difficult to use. Hall baby brain developmental approach, www.a-i.com, failed.www.a-i.com Open Mind Common Sense Project (MIT): a WWW collaboration with over 14,000 authors, who contributed 710,000 sentences; used to generate ConceptNet, very large semantic network. Some interesting projects are being developed now around this network but no systematic knowledge has been collected. Other such projects: HowNet (Chinese Academy of Science), FrameNet (Berkeley), various large-scale ontologies, MindNet (Microsoft) project, to improve translation. Mostly focused on understanding of all relations the in text/dialogue.
- Slide 16
- Challenges: language Turing test original test is too difficult. Loebner Prize competition, for almost two decades played by chatterbots based on template or contextual pattern matching cheating can get you quite far... A personal Turing test (Carpenter and Freeman), with programs trying to impersonate real personally known individuals. Question/answer systems; Text Retrieval Conf. (TREC) competitions. Word games, 20-questions game - knowledge of objects/properties, but not about complex relations between objects. Success in learning language depends on automatic creation, maintenance and the ability to use large- scale knowledge bases. Intelligent tutoring systems? How to define milestones?
- Slide 17
- Challenges: reasoning Super-expert system in a narrow domain (Feigenbaum), needs a lot of general intelligence to communicate, should reason in math, bioscience or law, experts will pose problems, probe understanding. Same direction, but without NLP: Automated Theorem Proving (ATM) System Competitions (CASC) in many sub-categories. General AI in math: general theorem provers, perhaps using meta-learning techniques with specialized modules + NLP. Automatic curation of genomic/pathways databases, creation of models of genetic and metabolic processes for bioorganisms. Partners that advice humans in their work, evaluating their reasoning (theorem checking), adding creative ideas, interesting associations.
- Slide 18
- Real AI? General purpose systems that can be taught skills needed to perform human jobs, and to measure which fraction of these jobs can be done by AI systems (Nilsson, Turings child machine). Knowledge-based information processing jobs progress measured by passing a series of examinations, ex. accounting. Manual labor requires senso-motoric coordination, harder to do? DARPA Desert & Urban Challenge competitions (2005/07), old technology, integrat
View more >