Evidence Dynamics in Neighborhood Models Johan van Benthem University of Amsterdam.

Slides:



Advertisements
Similar presentations
Hyperintensionality and Impossible Worlds: An Introduction
Advertisements

Formal Criteria for Evaluating Arguments
Meaningful learning.
Presentation on Artificial Intelligence
Computer Science CPSC 322 Lecture 25 Top Down Proof Procedure (Ch 5.2.2)
The Logic of Intelligence Pei Wang Department of Computer and Information Sciences Temple University.
Agents That Reason Logically Copyright, 1996 © Dale Carnegie & Associates, Inc. Chapter 7 Spring 2004.
Of 27 lecture 7: owl - introduction. of 27 ece 627, winter ‘132 OWL a glimpse OWL – Web Ontology Language describes classes, properties and relations.
INTERNATIONAL CONFERENCE ON NATURAL LANGUAGE PROCESSING NLP-AI IIIT-Hyderabad CIIL, Mysore ICON DECEMBER, 2003.
Introductory Lecture. What is Discrete Mathematics? Discrete mathematics is the part of mathematics devoted to the study of discrete (as opposed to continuous)
CPSC 322, Lecture 19Slide 1 Propositional Logic Intro, Syntax Computer Science cpsc322, Lecture 19 (Textbook Chpt ) February, 23, 2009.
Dutch books and epistemic events Jan-Willem Romeijn Psychological Methods University of Amsterdam ILLC 2005 Interfacing Probabilistic and Epistemic Update.
Peter Gärdenfors From communication to logic. From logic to communication Classical logic: Syllogisms Propositional logic Predicate logic Reduction of.
Meaning and Language Part 1.
Foundations This chapter lays down the fundamental ideas and choices on which our approach is based. First, it identifies the needs of architects in the.
UNIT 9. CLIL THINKING SKILLS
Module 1: A Closer Look at the Common Core State Standards for Mathematics High School Session 2: Matching Clusters of Standards to Critical Areas in one.
1 Unit 4: One-Step Equations The Georgia Performance Standards Website.
Teaching Teaching Discrete Mathematics and Algorithms & Data Structures Online G.MirkowskaPJIIT.
CSNB143 – Discrete Structure
‘Pure Mathematics is, in its way, the poetry of logical ideas’ Einstein ‘Maths is like love, a simple idea but it can get very complicated.’ Unknown ‘The.
1 Mathematical Institute Serbian Academy of Sciences and Arts, Belgrade DEUKS Meeting Valencia, September 9-11, 2008, Valencia New PhD modules proposal.
Applying Belief Change to Ontology Evolution PhD Student Computer Science Department University of Crete Giorgos Flouris Research Assistant.
Easy steps to writing THE ESSAY. Writing an essay means: Creating ideas from information Creating arguments from ideas Creating academic discourse to.
CLC reading program Nguyen Thi Thu Trang. In-class activities Assignment Assessment Add your text in here Reading program Objectives Contents.
1 LIS590IM Information Modeling — Slide Set for Class 16 The Father Guido Sarducci Slide and some final comments Slides for Dec 16 lecture LIS590IML: Information.
The tasks of logic Why we need more versatile tools Philosophy and logic 2013 Kyiv 25 May
Fall 98 Introduction to Artificial Intelligence LECTURE 7: Knowledge Representation and Logic Motivation Knowledge bases and inferences Logic as a representation.
Discrete Structures for Computing
LOGIC AND ONTOLOGY Both logic and ontology are important areas of philosophy covering large, diverse, and active research projects. These two areas overlap.
Knowledge, Representation, and Reasoning CSC 244/444: Logical Foundations of Artificial Intelligence Henry Kautz.
Epistemic Strategies and Games on Concurrent Processes Prakash Panangaden: Oxford University (on leave from McGill University). Joint work with Sophia.
Copyright © Curt Hill Mathematical Logic An Introduction.
Designing software architectures to achieve quality attribute requirements F. Bachmann, L. Bass, M. Klein and C. Shelton IEE Proceedings Software Tzu-Chin.
Semantic web course – Computer Engineering Department – Sharif Univ. of Technology – Fall Knowledge Representation Semantic Web - Fall 2005 Computer.
A Common Ground for Virtual Humans: Using an Ontology in a Natural Language Oriented Virtual Human Architecture Arno Hartholt (ICT), Thomas Russ (ISI),
Research Here and There Zhizheng Zhang( 张志政 ) Southeast University( 东南大学 )
School of Computing and Mathematics, University of Huddersfield CHA2545: WEEK 4 LECTURE: DENOTIONAL SEMANTICS OF A SIMPLE LANGUAGE TUTORIAL: Do exercises.
Acknowledgements This project would not have been possible without the gracious help of the McNair Scholars Program. The authors of this project thank.
Of 33 lecture 1: introduction. of 33 the semantic web vision today’s web (1) web content – for human consumption (no structural information) people search.
Basic Concepts of Logic An Overview of Introduction to Logic Yingrui Yang
1 Viewing Vision-Language Integration as a Double-Grounding case Katerina Pastra Department of Computer Science, Natural Language Processing Group, University.
Issues in Ontology-based Information integration By Zhan Cui, Dean Jones and Paul O’Brien.
THE IMPORTANCE OF DISCRETE MATHEMATICS IN COMPUTER TECHNOLOGY.
From Hoare Logic to Matching Logic Reachability Grigore Rosu and Andrei Stefanescu University of Illinois, USA.
Computer Science CPSC 322 Lecture 22 Logical Consequences, Proof Procedures (Ch 5.2.2)
Intertheoretic Reduction and Explanation in Mathematics
International Conference on Fuzzy Systems and Knowledge Discovery, p.p ,July 2011.
ACE TESOL Diploma Program – London Language Institute OBJECTIVES You will understand: 1. A variety of interactive techniques that cater specifically to.
Formal Specification: a Roadmap Axel van Lamsweerde published on ICSE (International Conference on Software Engineering) Jing Ai 10/28/2003.
BC Curriculum Revisions 1968 → what 1976 → what 1984 → what + how 1994 → what + how 2003 → what + how 2008 → what + how 2015 → how + what.
Chapter 1 What is Biology? 1.1 Science and the Natural World.
Epistemology: Theory of Knowledge Question to consider: What is the most reliable method of knowing?
WonderWeb. Ontology Infrastructure for the Semantic Web. IST WP4: Ontology Engineering Heiner Stuckenschmidt, Michel Klein Vrije Universiteit.
Intelligent Agents Chapter 2. How do you design an intelligent agent? Definition: An intelligent agent perceives its environment via sensors and acts.
5 Lecture in math Predicates Induction Combinatorics.
Introductory Lecture. What is Discrete Mathematics? Discrete mathematics is the part of mathematics devoted to the study of discrete (as opposed to continuous)
UEF // University of Eastern Finland How to publish scientific journal articles? 10 STEPS TO SUCCESS lllllllllllllllllllllllllllllllll lllllllllllllllllllllllllllllllll.
Artificial Intelligence Logical Agents Chapter 7.
WHAT MODELS DO THAT THEORIES CAN’T Lilia Gurova Department of Cognitive Science and Psychology New Bulgarian University.
A Normative and Intentional Agent Model for Organisation Modelling
Exploring the relation between knowledge and belief in multi-agent epistemic planning. Number 3.
Logical Agents.
Intelligent Agents Chapter 2.
Discrete Mathematics and Its Applications
EPISTEMIC LOGIC.
ece 720 intelligent web: ontology and beyond
Introduction to course
Ontology.
ONTOMERGE Ontology translations by merging ontologies Paper: Ontology Translation on the Semantic Web by Dejing Dou, Drew McDermott and Peishen Qi 2003.
Presentation transcript:

Evidence Dynamics in Neighborhood Models Johan van Benthem University of Amsterdam & Stanford University István Németi 70 Conference, Rényi Institute, Budapest, 8 Sept 2012

Current Sources

Evidence for Beliefs Rational beliefs are grounded in reasons Double aspect: ‘ support ’, and ‘ hooks for refutation ’ Many views of reasons or evidence in epistemology We will pursue only an austere view of evidence, that will still lead us to lush logical pastures.

In Between Syntax and Semantics The barest view: one ’ s current evidence is the range of all epistemically accessible worlds. Ignores how we got here. The richest view: all the details of what we learnt so far, including current syntactic theory. In between: family of subsets of the domain encoding evidence from various sources, at an abstract level. Not always consistent: sources may be unreliable.

Ubiquitous Idea Belief revision Modeling theories in philosophy of science Premise semantics for conditionals Fine-grained views of knowledge Knowledge obtained through sensors

Other Important Approaches

Models for Evidence p

Neighborhood Semantics

Language: Evidence, Belief, Knowledge Language: Evidence, Belief, Knowledge

Truth Conditions

First Stab: Belief is Cautious

Static Base Logic Problem Axiomatize this logic: simple + simple ≠ simple?

One Static Logic of Evidence Models Joint work with David Fernandez: Proof uses quasi-models, representation, NBD bisimulation.

Designing a Better Language Two different sources for choosing logical formalisms: The needs of existing discourse, reasoning scenarios What is suggested by the semantic models themselves Issue in philosophy and cognitive science: What is our repertoire of epistemic attitudes? For instance, belief seems not one notion, but many.

First: Conditional Evidence

Adding Conditional Belief Classic addition needed in many places (though curiously overlooked in doxastic logic, belief revision theory):

Base Logic Extended

Logic of Conditional Belief Is more or less the ‘minimal conditional logic’:

Dynamics of Evidence Change Evidence comes in a stream, with all the events that we know from logical dynamics in general. Main purpose: explore more fine-grained information dynamics living at the level of neighborhood models. Usual tool: recursion axioms as key ‘ dynamic equations ’. But also new angle on DEL: explore optimal design of static base language by dynamic-epistemic techniques.

Evidence Change: Announcement Hard information by public announcement

Recursion Axioms for Announcement

Deconstructing Public Announcement

Evidence Change by Addition

Recursion Axioms as Design Tool Issue: how much of such ‘ culture ’ in a course?

Graphical Pictures Can Help Issue: how much of such ‘ culture ’ in a course?

Conditioning Comes in Varieties Issue: how much of such ‘ culture ’ in a course?

New Logical Distinctions Issue: how much of such ‘ culture ’ in a course?

Enriched Base Language One dynamic task leads to another:

Discarding as an Action Discarding or retraction has seemed a challenge to DEL:

Extending the Language Once More Issue: how much of such ‘ culture ’ in a course?

Recursion Axioms for Removal Issue: how much of such ‘ culture ’ in a course?

The Complete Dynamic Logic of Removal Issue: how much of such ‘ culture ’ in a course?

Internal Operations on Evidence What about new operations without earlier counterparts?

Evidence Combination Issue: how much of such ‘ culture ’ in a course?

Logic of Evidence Combination Issue: how much of such ‘ culture ’ in a course?

The Complete Static Logic?

‘ Second Opinion ’ : Plausibility Models The other main approach to belief: add a plausibility order to epistemic models. How does this relate?

Parallels: Enriched Languages Richer languages: Levesque, Stalnaker, Battigalli & Sinischalci

Semantic Rock Bottom Issue: how much of such ‘ culture ’ in a course?

Not Quite Follow Received Wisdom Issue: how much of such ‘ culture ’ in a course?

Crossing Over: Representation Issue: how much of such ‘ culture ’ in a course?

An Illustration Issue: how much of such ‘ culture ’ in a course?

Translation Between Languages Issue: how much of such ‘ culture ’ in a course?

From Evidence to Plausibility Issue: how much of such ‘ culture ’ in a course?

Illustrations

The Total Picture Issue: how much of such ‘ culture ’ in a course?

Partial Translation Issue: how much of such ‘ culture ’ in a course?

Extends to Dynamic Parallels Through the earlier translations, Evidence addition becomes the relational operation called suggest(  ) changing the current order ≤ to its subrelation where no  -world is preferred to a ¬  -world. But evidence removal has no such relational counterpart. The full extent of harmony between the neighborhood/ evidence and ordering levels remains to be understood. Technical questions remain (cf. Parikh ’ s trick for dynamic game logic): mimick all of the NBD via extra relations?

Interim Conclusion Plausibility models a special case of neighborhood models. They, too, support many static notions of belief, and the usual DEL plausibility update mechanism can analyze their dynamics. But neighborhood models also allow for internal operations, and seem the more fine-grained perspective eventually. Technical questions remain: mimick NBD via extra relations?

Technical Challenges Complete axiomatizations new neighborhood logics. New kinds of bisimulation as expressive power varies. Clarify connections to modal logics of plausibility order. For the algebraists How to deal with dynamic operators that change models? From classical to intuitionistic base, from distributive to just a monotonic base algebra? (Ma, Palmigiano & Sadrzadeh 2011, Palmigiano & Kurz 2012.)

Richer Views of Evidence World dependence and iterated evidence Many agents, with new operations (‘merge’) Resolving conflicts in the current evidence Trust, priority, and authority Manipulating reasons Richer syntactic views than neighborhoods Weiging evidence and numerical approaches.

Conclusion Evidence is a key notion worthy of logical study. From trick, neighborhood models become avant garde topic when given a language of their own. Poses new mathematical issues for modal logics (and perhaps also for algebraic logic). Especially when combined with logical dynamics.