scholarly journals Improving First-Order Optimization Algorithms (Student Abstract)

2020 ◽  
Vol 34 (10) ◽  
pp. 13935-13936
Author(s):  
Tato Ange ◽  
Nkambou Roger
Keyword(s):  
Optimization Algorithms ◽  
Rule Based ◽  
First Order ◽  
Previous Step

This paper presents a simple and intuitive technique to accelerate the convergence of first-order optimization algorithms. The proposed solution modifies the update rule, based on the variation of the direction of the gradient and the previous step taken during training. Results after tests show that the technique has the potential to significantly improve the performance of existing first-order optimization algorithms.

scholarly journals Cautious Rule-Based Collective Inference

2019 ◽  
Author(s):  
Martin Svatos
Keyword(s):  
Relational Learning ◽  
Statistical Relational Learning ◽  
Knowledge Graph ◽  
Inference Process ◽  
Rule Based ◽  
Test Error ◽  
Popular Approach ◽  
First Order ◽  
Theoretical Test ◽  
Collective Inference

Collective inference is a popular approach for solving tasks as knowledge graph completion within the statistical relational learning field. There are many existing solutions for this task, however, each of them is subjected to some limitation, either by restriction to only some learning settings, lacking interpretability of the model or theoretical test error bounds. We propose an approach based on cautious inference process which uses first-order rules and provides PAC-style bounds.

scholarly journals Reasoning About Loops Using Vampire

10.29007/tcvj ◽  
2018 ◽  
Author(s):  
Laura Kovács ◽  
Simon Robillard
Keyword(s):  
Order Logic ◽  
Theorem Prover ◽  
Program Correctness ◽  
Command Language ◽  
Rule Based ◽  
Invariant Generation ◽  
Partial Correctness ◽  
First Order ◽  
Deductive Systems ◽  
Fully Automatic

In 2009, the symbol elimination method for loop invariant generationwas introduced, which used saturationtheorem proving in first-order logic to generate quantified invariantsof programs with arrays. Symbol elimination is fully automatic,requires no user guidance, and it is the first ever approach able togenerate invariants with alternations of quantifiers. In this paperwe describe a number of improvements and extensions to symbolelimination and invariant generation using first-order theoremproving, in particular the Vampire theorem prover. Rather than beinglimited to a specific programming language, our approach to reasoningabout loops in Vampire relies on a simple guarded command language forits input, which can be used as an interface for more complex andrealistic imperative languages. We propose new ways for extendingquantified loop properties describing valid loop properties, bysimplifying the properties over array updates and next staterelations. We also extend symbol elimination with pre- andpost-conditions of loops. We use the loop specification to generateonly invariants that are relevant, that is, invariants that are neededfor proving partial correctness of loops. Further, we turn symbolelimination into an automatic approach proving program correctness,providing an alternative method to Hoare-rule based loop verificationor other deductive systems. We present our newly redesignedimplementation of loop reasoning in Vampire and also report onexperimental results.

Accelerated First-Order Optimization Algorithms for Machine Learning

2020 ◽  
Vol 108 (11) ◽  
pp. 2067-2082
Author(s):  
Huan Li ◽  
Cong Fang ◽  
Zhouchen Lin
Keyword(s):  
Machine Learning ◽  
Optimization Algorithms ◽  
First Order

Towards a Specification Language for Mobile Applications

2013 ◽  
Vol 5 (2) ◽  
pp. 58-76
Author(s):  
Abdesselam Redouane
Keyword(s):  
Mobile Applications ◽  
Formal Language ◽  
Specification Language ◽  
Predicate Calculus ◽  
Rule Based ◽  
First Order ◽  
Logical Rule ◽  
Order Predicate Calculus

Mobile applications are increasingly being developed by many developers using different environments for diverse devices. However, there is no method or formal language that helps these developers specify their requirements before the coding stage. In this paper, the author describes a specification language that helps in this context. To achieve rigor, the language is based on an extended first order predicate calculus and allows specifiers to modularize their specifications. The language is founded on two concepts: a pristine which can be seen as a unit of cognition and a definition which is a logical rule based on pristines and other definitions. Furthermore, the language allows specifiers to express constraints which are fundamental features in mobile applications. A use of this language is shown by an example.

Principles and practice in verifying rule-based systems

1992 ◽  
Vol 7 (2) ◽  
pp. 115-141 ◽  
Author(s):  
Alun D. Preece ◽  
Rajjan Shinghal ◽  
Aïda Batarekh
Keyword(s):  
Expert System ◽  
Knowledge Base ◽  
State Of The Art ◽  
Knowledge Bases ◽  
Order Logic ◽  
First Order Logic ◽  
Rule Based ◽  
First Order ◽  
Rule Bases ◽  
System Knowledge

AbstractThis paper surveys the verification of expert system knowledge bases by detecting anomalies. Such anomalies are highly indicative of errors in the knowledge base. The paper is in two parts. The first part describes four types of anomaly: redundancy, ambivalence, circularity, and deficiency. We consider rule bases which are based on first-order logic, and explain the anomalies in terms of the syntax and semantics of logic. The second part presents a review of five programs which have been built to detect various subsets of the anomalies. The four anomalies provide a framework for comparing the capabilities of the five tools, and we highlight the strengths and weaknesses of each approach. This paper therefore provides not only a set of underlying principles for performing knowledge base verification through anomaly detection, but also a survey of the state-of-the-art in building practical tools for carrying out such verification. The reader of this paper is expected to be familiar with first-order logic.

Sign in / Sign up

Export Citation Format

Share Document