Self-referencing Agents for Inductive Non-Algorithmic e-Learning
نویسندگان
چکیده
The paper presents an alternative approach to e-Learning, where “Learning” is action-oriented and highly personalised, while “e-” is carried out through a software entity acting as self-referencing coach and interacting with the user as interface agent. The focus is on aspects related to: a) domain theory (moving targets, refined strategies); b) trends in computer science (uncertain knowledge processing, obsolescence of algorithmic programming, and above all new ontologies); c) affordability (upholding a software engineering perspective, mainly an agent-oriented one). Specific objectives are: a) To defend the rationale for an unconventional outlook about e-Learning pertaining to all main system aspects. b) To draw a stepwise approach affordable within a limited academic research. c) To outline a generic architecture for successive experimental models and to present very roughly the current one. d) To employ this project as test field for a more comprehensive undertaking. (Details and implementation issues regarding mechanisms and models are described in other papers.) Preliminary estimations are encouraging as regards perspective, methods, and generic architecture. Key-Words: Non-algorithmic Learning; Anthropocentric Interface; Induction; Self-Cloning Agent; Ontology.
منابع مشابه
DESIGNING A CURRICULUM MODEL FOR GENERAL MEDICINE WITH A COMBINED METHOD (E-LEARNING AND NON-E-LEARNING) INSPIRED BY THE AKKER MODEL: A QUALITATIVE STUDY
Background & Aims: In addition to providing health care services, medical universities have an important role in training expert and skilled manpower needed by different sections of society. In order to do so, the general medical education curriculum should be constantly reviewed and improved by eliminating the shortcomings. The aim of this study was to design a curriculum model for teaching ge...
متن کاملUltimate Intelligence Part II: Physical Measure and Complexity of Intelligence
We investigate physical measures and limits of intelligence that are objective and useful. We propose a universal measure of operator induction fitness, and show how it can be used in a reinforcement learning model, and a self-preserving agent model based on the free energy principle. We extend logical depth and conceptual jump size measures to stochastic domains, and elucidate their relations....
متن کاملOptimal Language Learning
Gold’s original paper on inductive inference introduced a notion of an optimal learner . Intuitively, a learner identifies a class of objects optimally iff there is no other learner that: requires as little of each presentation of each object in the class in order to identify that object, and, for some presentation of some object in the class, requires less of that presentation in order to iden...
متن کاملGeneral Limitations on Machine Learning
Machine learning is widely regarded as a tool for overcoming the bottleneck in knowledge acquisition. Especially in knowledge-intensive domains there is the hope for using machine learning techniques successfully. This paper prove the general inability of simple learning programs to learn complex concepts from few input data. This holds independently of the epistemological problems of inductive...
متن کاملSafe learning agents? Logic-based agents!
Safe learning agents are agents whose learned behaviour can be predicted and analysed. Non-symbolic learning algorithms such as reinforcement learning rely on emergence and thus are not a good candidate to building safe AI systems. Our contention is that logicbased algorithms such as explanation-based learning and inductive logic programming should instead be used to design and implement safe, ...
متن کامل