Autonomous Machine Learning What kind of a-priori knowledge do we have to provide to our systems for showing such a capability? Edgar Koerner Honda Research.

Slides:



Advertisements
Similar presentations
E-Learning E-learning means all teaching and learning forms supported by modern ICT. The e-content can be organized interactively and in a multimedia.
Advertisements

Pat Langley School of Computing and Informatics Arizona State University Tempe, Arizona USA Modeling Social Cognition in a Unified Cognitive Architecture.
Modelling with expert systems. Expert systems Modelling with expert systems Coaching modelling with expert systems Advantages and limitations of modelling.
TU/e technische universiteit eindhoven Hera: Development of Semantic Web Information Systems Geert-Jan Houben Peter Barna Flavius Frasincar Richard Vdovjak.
TeachMed Clinical reasoning learning with simulated patients Guy Bisson M.D., Faculté de médecine Froduald Kabanza Ph.D., Faculté des sciences, département.
How the brain relates inputs to conclude in an output/outputs Mostafa M. Dini.
Chapter 11 user support. Issues –different types of support at different times –implementation and presentation both important –all need careful design.
15 th International Conference on Design Theory and Methodology 2-6 September 2003, Chicago, Illinois Intelligent Agents in Design Zbigniew Skolicki Tomasz.
Lecture 1 Introduction. General Theory of Foreign Languages Teaching and Learning (FLTL)
Presented by: Thabet Kacem Spring Outline Contributions Introduction Proposed Approach Related Work Reconception of ADLs XTEAM Tool Chain Discussion.
COGNITIVE VIEWS OF LEARNING Information processing is a cognitive theory that examines the way knowledge enters and is stored in and retrieved from memory.
Employability in context of the Bologna Process Gayane Harutyunyan Bologna Secretariat Yerevan, May 2014.
Multiple V-model. Introduction In embedded systems, the test object is not just executable code. First a model of the system is built on a PC, which simulates.
Yiannis Demiris and Anthony Dearden By James Gilbert.
TU/e technische universiteit eindhoven Hypermedia Presentation Adaptation on the Semantic Web Flavius Frasincar Geert-Jan Houben
King Saud University College of nursing Master program.
Dealing with Complexity Peter Andras Department of Psychology University of Newcastle
The Importance of Architecture for Achieving Human-level AI John Laird University of Michigan June 17, th Soar Workshop
Modeling and the simulator of Digital Circuits in Object-Oriented Programming Stefan Senczyna Department of Fundamentals of Technical Systems The Silesian.
Marakas: Decision Support Systems, 2nd Edition © 2003, Prentice-Hall Chapter Chapter 7: Expert Systems and Artificial Intelligence Decision Support.
4. Interaction Design Overview 4.1. Ergonomics 4.2. Designing complex interactive systems Situated design Collaborative design: a multidisciplinary.
Chapter 12: Intelligent Systems in Business
1 Presenter: Chien-Chih Chen Proceedings of the 2002 workshop on Memory system performance.
Cognitive Science Overview Ausubel’s Meaningful Reception Learning Theory Schema Theory Advance Organizers Managing Essential Processing Application to.
The Brain, Learning, and Memory Key: AWL to Study, Low-frequency Vocabulary What is the connection between the brain, learning, and memory?
Cognitive Psychology, 2 nd Ed. Chapter 1. Defining Cognitive Psychology The study of human mental processes and their role in thinking, feeling, and behaving.
Click to edit Master title style  Click to edit Master text styles  Second level  Third level  Fourth level  Fifth level  Click to edit Master text.
HOW TO SOLVE IT? Algorithms. An Algorithm An algorithm is any well-defined (computational) procedure that takes some value, or set of values, as input.
PROCESSING APPROACHES
Teaching Vocabulary.
Educational Services for Individuals with Exceptionalities Adapted Lesson Plan.
The role of hypermedia in developing different kinds of vocational knowledge Fred Beven Griffith University.
Business Analysis and Essential Competencies
 Knowledge Acquisition  Machine Learning. The transfer and transformation of potential problem solving expertise from some knowledge source to a program.
Memory. Interesting Video  Color Changing Card Trick Color Changing Card Trick.
Help Desk System How to Deploy them? Author: Stephen Grabowski.
Mobile Topic Maps for e-Learning John McDonald & Darina Dicheva Intelligent Information Systems Group Computer Science Department Winston-Salem State University,
MEMORY & INTELLIGENCE.
Towards Cognitive Robotics Biointelligence Laboratory School of Computer Science and Engineering Seoul National University Christian.
MRPGA : An Extension of MapReduce for Parallelizing Genetic Algorithm Reporter :古乃卉.
Development Process and Testing Tools for Content Standards OASIS Symposium: The Meaning of Interoperability May 9, 2006 Simon Frechette, NIST.
Sociocultural Theory Week 4, “Sociocultural Approaches to Learning and Development”
Chapter 10 Analysis and Design Discipline. 2 Purpose The purpose is to translate the requirements into a specification that describes how to implement.
Author: Gilbert Paquette Reuse freely – Just quote Meta-Knowledge Representation for Learning Systems (Part 1-What) Meta-Knowledge Representation for Learning.
Visual Information Systems Recognition and Classification.
MAPLD 2005/254C. Papachristou 1 Reconfigurable and Evolvable Hardware Fabric Chris Papachristou, Frank Wolff Robert Ewing Electrical Engineering & Computer.
Slide 1Reproduction prohibited without permission from Computas AS © METIS An Open Architecture Toolkit ADM and ADML support Don Hodge Principle Knowledge.
1 Text Reference: Warford. 2 Computer Architecture: The design of those aspects of a computer which are visible to the programmer. Architecture Organization.
Cognitive Theories of Learning Dr. K. A. Korb University of Jos.
World Representation for Vehicle Navigation and Standards for Cooperative Vehicles Dr Javier Ibanez-Guzman 31st, January 2007 Orbassano.
Chapter 10 Memory and Thought. The Processes of Memory The storage and retrieval of what has been learned or experienced is memory There are three processes.
MEMORY & INTELLIGENCE. MEMORY: The input, storage, and retrieval of what has been learned or experienced.
Chapter 7 Consumer Learning.
Towards Unifying Vector and Raster Data Models for Hybrid Spatial Regions Philip Dougherty.
RULES Patty Nordstrom Hien Nguyen. "Cognitive Skills are Realized by Production Rules"
Manufacturing Systems Integration Division Development Process and Testing Tools for Content Standards Simon Frechette National Institute of Standards.
Chapter 7 Memory. Objectives 7.1 Overview: What Is Memory? Explain how human memory differs from an objective video recording of events. 7.2 Constructing.
 Knowledge Acquisition  Machine Learning. The transfer and transformation of potential problem solving expertise from some knowledge source to a program.
3.1 Cognitive processes. Cognitive psychology Includes: perception, thinking, problem solving, memory, language, and attention. Cognition refers to such.
MODELS OF CONSUMER BEHAVIOUR
Memory Module One: Booklet #8.
A class room seminar on SCHEMA THEORY(BARTLETT)
Memory Module One: Booklet #8.
Introduction SWE 619.
Cognitive level of analysis
Linked List and Selection Sort
Chapter 11 user support.
Cognitive Level of Analysis: Cognitive Processes
LEARNING.
A.4 Innate and Learned Behavior
Presentation transcript:

Autonomous Machine Learning What kind of a-priori knowledge do we have to provide to our systems for showing such a capability? Edgar Koerner Honda Research Institute Europe Offenbach, Germany

WCCI Barcelona 2 Prerequisites for autonomous knowledge acquisition Autonomous learning, learning while behaving –Learner is the subject, defines by itself what to learn Requires subjective knowledge representation Behaviour provides semantics to data that the system is in a position to define what to learn and where in the system to learn Capability to associate current situation with acquired experience to define what is new, and in which relation to the already acquired knowledge it should be memorized –Living beings are still the only example of autonomous learners in complex environments Living beings without cortex: autonomous systems (reflex automatons) genetically encoded reflex hierarchy genetically encoded “value system” = mapping of sensory prototypic situations to behavioural prototypes Living beings with cortex: flexible autonomous systems (learning systems) reflex hierarchy value system + memory system (cortex) mapping sensory situation to behaviour control, predominantly genetically determined + self-referential control (also genetically encoded) to enable the system to generate a consistent relational architecture of the knowledge representation

WCCI Barcelona 3 Self-referential control is the key for autonomous learning Self-referential control may serve as a “representational immune system” An autonomous learning system must have the capability to decompose its sensory input according to its already acquired knowledge For an autonomous system, the only way to “understand” input is to relate it to already stored experience (association) If the complete input cannot be closely matched to existing experience, the “width of association” must be controlled in a way to enable both coarse sorting for behavioral categorization, and discrimination of the new aspects Activating the next nearest existing representation anchors the situation in the semantics of behavioural relevance Subsequent iterative decomposition of the sensory input by existing representations can create a cumulative extension linked at the anchor point for integration into the already acquired subjective knowledge representation Understanding the sensory input by decomposing it into existing representational elements and remaining new aspects (residual) ensures Consistent relational structure of knowledge representation Efficient representation Isolating the new → selective attention Incrementally building up and correcting knowledge representation Basis for internal simulation Active disambiguation in complex situations, etc.