An official website of the United States government
We consider the problem of answering queries about formulas of first-order logic based on background knowledge partially represented explicitly as other formulas, and partially represented as examples independently drawn from a fixed probability distribution. PAC semantics, introduced by Valiant, is one rigorous, general proposal for learning to reason in formal languages: although weaker than classical entailment, it allows for a powerful model theoretic framework for answering queries while requiring minimal assumptions about the form of the distribution in question. To date, however, the most significant limitation of that approach, and more generally most machine learning approaches with robustness guarantees, is that the logical language is ultimately essentially propositional, with finitely many atoms. Indeed, the theoretical findings on the learning of relational theories in such generality have been resoundingly negative. This is despite the fact that first-order logic is widely argued to be most appropriate for representing human knowledge. In this work, we present a new theoretical approach to robustly learning to reason in first-order logic, and consider universally quantified clauses over a countably infinite domain. Our results exploit symmetries exhibited by constants in the language, and generalize the notion of implicit learnability to show how queries can be computed against (implicitly) learned first-order background knowledge.
more »
« less
A Categorical Approach to DIBI Models
The logic of Dependence and Independence Bunched Implications (DIBI) is a logic to reason about conditional independence (CI); for instance, DIBI formulas can characterise CI in discrete probability distributions and in relational databases, using a probabilistic DIBI model and a similarly-constructed relational model. Despite the similarity of the two models, there lacks a uniform account. As a result, the laborious case-by-case verification of the frame conditions required for constructing new models hinders them from generalising the results to CI in other useful models such that continuous distribution. In this paper, we develop an abstract framework for systematically constructing DIBI models, using category theory as the unifying mathematical language. We show that DIBI models arise from arbitrary symmetric monoidal categories with copy-discard structure. In particular, we use string diagrams - a graphical presentation of monoidal categories - to give a uniform definition of the parallel composition and subkernel relation in DIBI models. Our approach not only generalises known models, but also yields new models of interest and reduces properties of DIBI models to structures in the underlying categories. Furthermore, our categorical framework enables a comparison between string diagrammatic approaches to CI in the literature and a logical notion of CI, defined in terms of the satisfaction of specific DIBI formulas. We show that the logical notion is an extension of string diagrammatic CI under reasonable conditions.
more »
« less
- Award ID(s):
- 2153916
- PAR ID:
- 10574790
- Editor(s):
- Rehof, Jakob
- Publisher / Repository:
- Schloss Dagstuhl – Leibniz-Zentrum für Informatik
- Date Published:
- Volume:
- 299
- ISSN:
- 1868-8969
- ISBN:
- 978-3-95977-323-2
- Page Range / eLocation ID:
- 299-299
- Subject(s) / Keyword(s):
- Conditional Independence Dependence Independence Bunched Implications String Diagrams Markov Categories Theory of computation → Logic Theory of computation → Semantics and reasoning Theory of computation → Models of computation
- Format(s):
- Medium: X Size: 20 pages; 1014792 bytes Other: application/pdf
- Size(s):
- 20 pages 1014792 bytes
- Right(s):
- Creative Commons Attribution 4.0 International license; info:eu-repo/semantics/openAccess
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Conditional independence (CI) tests play a central role in statistical inference, machine learning, and causal discovery. Most existing CI tests assume that the samples are indepen- dently and identically distributed (i.i.d.). How- ever, this assumption often does not hold in the case of relational data. We define Relational Conditional Independence (RCI), a generaliza- tion of CI to the relational setting. We show how, under a set of structural assumptions, we can test for RCI by reducing the task of test- ing for RCI on non-i.i.d. data to the problem of testing for CI on several data sets each of which consists of i.i.d. samples. We develop Kernel Relational CI test (KRCIT), a nonpara- metric test as a practical approach to testing for RCI by relaxing the structural assumptions used in our analysis of RCI. We describe re- sults of experiments with synthetic relational data that show the benefits of KRCIT relative to traditional CI tests that don’t account for the non-i.i.d. nature of relational data.more » « less
-
null (Ed.)We investigate constructions of higher arity self-distributive operations, and give relations between cohomology groups corresponding to operations of different arities. For this purpose we introduce the notion of mutually distributive [Formula: see text]-ary operations generalizing those for the binary case, and define a cohomology theory labeled by these operations. A geometric interpretation in terms of framed links is described, with the scope of providing algebraic background of constructing [Formula: see text]-cocycles for framed link invariants. This theory is also studied in the context of symmetric monoidal categories. Examples from Lie algebras, coalgebras and Hopf algebras are given.more » « less
-
The linear decomposition attack provides a serious obstacle to direct applications of noncommutative groups and monoids (or semigroups) in cryptography. To overcome this issue we propose to look at monoids with only big representations, in the sense made precise in the paper, and undertake a systematic study of such monoids. One of our main tools is Green’s theory of cells (Green’s relations). A large supply of monoids is delivered by monoidal categories. We consider simple examples of monoidal categories of diagrammatic origin, including the Temperley–Lieb, the Brauer and partition categories, and discuss lower bounds for their representations.more » « less
-
null (Ed.)Robust Linear Temporal Logic (rLTL) was crafted to incorporate the notion of robustness into Linear-time Temporal Logic (LTL) specifications. Technically, robustness was formalized in the logic rLTL via 5 different truth values and it led to an increase in the time complexity of the associated model checking problem. In general, model checking an rLTL formula relies on constructing a generalized Büchi automaton of size 5^|φ| where |φ| denotes the length of an rLTL formula φ. It was recently shown that the size of this automaton can be reduced to 3^|φ| (and even smaller) when the formulas to be model checked come from a fragment of rLTL. In this paper, we introduce Evrostos, the first tool for model checking formulas in this fragment. We also present several empirical studies, based on models and LTL formulas reported in the literature, confirming that rLTL model checking for the aforementioned fragment incurs in a time overhead that makes the verification of rLTL practical.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.4230/LIPIcs.FSCD.2024.17
