140 likes | 255 Views
On the Generalized Deduction, Induction and Abduction as the Elementary Reasoning Operators within Computational Semiotics. Faculty of Electrical and Computer Engineering State University of Campinas FEEC - UNICAMP - Brazil. Ricardo R. Gudwin. Introduction.
E N D
On the Generalized Deduction, Induction and Abduction as the Elementary Reasoning Operators within Computational Semiotics Faculty of Electrical and Computer Engineering State University of Campinas FEEC - UNICAMP - Brazil Ricardo R. Gudwin
Introduction • Computational Semiotics - attempt of emulating the semiosis cycle within a digital computer • Intelligent Behavior semiotic processing within an autonomous system • Intelligent System Semiotic System • Key issue : • discovery of elementary/minimum units of intelligence relation to Semiotics • Current Efforts: • Albus’ Outline for a Theory of Intelligence • Meystel’s GFACS algorithm • Alternative Set of Operators: • knowledge extraction (abstraction for deduction) • knowledge generation (abstraction for induction) • knowledge selection (abstraction for abduction)
Knowledge Units • Duality :Information x Knowledge (what’s the difference ?) • Knowledge Unit : “A granule of information encoded into a structure” • How does a system obtain knowledge units ? • Environment - • set of dynamical continuous phenomena running in parallel • cannot be known as a whole • Sensors - • provide a partial and continuous source of information • Umwelt (Uexkull, 1986) - sensible environment • How to encode such information into knowledge ? • Singularities Extraction knowledge units
Knowledge Units • Singularities • discrete entities that model, in a specific level of resolution, phenomena occurring in the world • need to be encoded to become knowledge units • Codification • representation space • embodiment vehicle (structure) • Structures • numbers • lists • trees • graphs
Knowledge Units • Representation Space • after interpretation • before interpretation : focus of attention mechanism
Knowledge Units • Interpretation Problems: • structural identification problem • semantic identification problem • icon - data represents a direct model of phenomenon • index - data points to a localization within representation space where it is stored the direct model of phenomenon • symbol - data is only a key to be used in a conversion table (an auxiliary structure) that points to the direct model of phenomenon
Knowledge Units • Formation of Knowledge Units • Elementary Knowledge Units • singularity extraction mechanisms • More elaborate Knowledge Units • application of knowledge processing operators • A Taxonomy for Knowledge Units RIcObSp RIcSeG Sensors RIcObG RIn RSy DSy DIc RIcSeSp Actuator
S = { , , , , , ) S = { } = { , , , , , ) Packing Knowledge • Abstraction partial order relation ( ) • ab - b is an abstraction of a • extensional definition: • nominate each particular element belonging to a set • good for finite sets only • intensional definition: • define a set as the collection of all possible elements satisfying a condition • good for infinite sets • requires an encoding/decoding in order to convert from intensional to extensional representations • Examples: • S = {(x,y) R2 | y = 2x3+7x+1 } • S can be encoded by b = (2,0,7,1) • a = (1,10) , b = (2,0,7,1) ab • c = (0,1,1,10,2,31) T = {(0,1),(1,10),(2,31)} cb • a c b
Knowledge Extraction • P - Set of Premises • C - Set of Conclusions • C P • The blue knowledge units in P correspond to a packing of various red knowledge units • Obtaining C corresponds to the extraction of such knowledge units, compressed into P’s blue units
Knowledge Generation • P - Set of Premises • C - Set of Conclusions • P C • Obtaining C corresponds to the generation of new knowledge, using knowledge in P as a seed • This generation can happen by different ways: • combination, • fusion, • transformation (including insertion of noise, mutation, etc) • interpolation, • fitting, • topologic expansion
Knowledge Selection • P - Set of Premises • C - Set of Conclusions • H - Set of Hypothesis • C P • Obtaining C corresponds to a selection among candidates in H, using elements in P as a criteria • Elements in H can be obtained by any way: by a prior knowledge generation, randomly, etc.
Knowledge Operators xReasoning Operators • Similarity between knowledge operators and classical reasoning operators (deduction, induction, abduction) • Knowledge Extraction Generalized Deduction • Deduction : normally applied within logic (dicent knowledge units) • KE extends it to all types of knowledge units • Knowledge Generation Generalized Induction • Induction : process of producing a general proposition on the ground of a limited number of particular propositions • KG is more than induction. Induction is only one of KG procedures. KG includes operations (e.g. crossover, mutation) that are not usually categorized as induction • Knowledge Selection Generalized Abduction • The process of abduction can be decomposed into many phases: • anomaly detection deduction • explanatory hypothesis construction generalized induction • hypothesis verification • selection of best hypothesis generalized abduction
Building Intelligent Systems • Knowledge Units Mathematical Objects • Argumentative Knowledge Units Active Objects • Intelligent Systems Object Networks • Intelligent System for an AGV
Conclusions • GFACS and argumentative knowledge • Grouping generalized induction • Focusing Attention generalized deduction • Combinatorial Search generalized induction and abduction • Final Conclusions • Formalization of important issues regarding the intersection of semiotics and intelligent systems • Identification of three knowledge operators that are “atomic” for any type of intelligent system development • Foundations for a computational implementation of the semiosis loop under artificial systems • Background for the construction for intelligent systems theory, enhanced and sustained by computational semiotics