SPIN is an efficient verification system for models of distributed software systems. It has been used to detect design errors in applications ranging from high-level descriptions of distributed algorithms to detailed code for controlling telephone exchanges. The paper gives an overview of the design and structure of the verifier, reviews its theoretical foundation, and gives an overview of significant practical applications.

/pdf/the-model-checker-spin-1bgo4jtefp.pdf

The model checker SPIN

Explanation in artificial intelligence: Insights from the social sciences

http://emamispnu.persiangig.com/Projects/Cloud%20Computing/science_2.pdf

Review: A survey on security issues in service delivery models of cloud computing

Our goal in this paper is to introduce and motivate a methodology, called Tropos,1 for building agent oriented software systems. Tropos is based on two key ideas. First, the notion of agent and all related mentalistic notions (for instance goals and plans) are used in all phases of software development, from early analysis down to the actual implementation. Second, Tropos covers also the very early phases of requirements analysis, thus allowing for a deeper understanding of the environment where the software must operate, and of the kind of interactions that should occur between software and human agents. The methodology is illustrated with the help of a case study. The Tropos language for conceptual modeling is formalized in a metamodel described with a set of UML class diagrams.

/pdf/tropos-an-agent-oriented-software-development-methodology-4cfpp4zfvj.pdf

Tropos: An Agent-Oriented Software Development Methodology

Over the Internet today, computing and communications environments are significantly more complex and chaotic than classical distributed systems, lacking any centralized organization or hierarchical control. There has been much interest in emerging Peer-to-Peer (P2P) network overlays because they provide a good substrate for creating large-scale data sharing, content distribution, and application-level multicast applications. These P2P overlay networks attempt to provide a long list of features, such as: selection of nearby peers, redundant storage, efficient search/location of data items, data permanence or guarantees, hierarchical naming, trust and authentication, and anonymity. P2P networks potentially offer an efficient routing architecture that is self-organizing, massively scalable, and robust in the wide-area, combining fault tolerance, load balancing, and explicit notion of locality. In this article we present a survey and comparison of various Structured and Unstructured P2P overlay networks. We categorize the various schemes into these two groups in the design spectrum, and discuss the application-level network performance of each group.

/pdf/a-survey-and-comparison-of-peer-to-peer-overlay-network-cpcv2ij1hk.pdf

A survey and comparison of peer-to-peer overlay network schemes

In this paper a hierarchical structure of highly specialized decision procedures for subclasses of FOL of increasing complexity is presented. Each decision procedure is specialized on a recursive class of formulae and it can be exploited in an effective way by higher level procedures whose design and implementation is thus greatly simplified. The overall structure turns out to be highly modular, extendible and allows to face the decision problem at the “right level” of detail. This results in an architecture of well integrated and very efficient decision procedures.

From Propositional Deciders to First Order Deciders: a Structured Approach to the Decision Problem

/pdf/online-learning-of-action-models-for-pddl-planning-3canpwe8t2.pdf

Online Learning of Action Models for PDDL Planning

Autonomous agents embedded in a physical environment need the ability to recognize objects and their properties from sensory data. Such a perceptual ability is often implemented by supervised machine learning models, which are pre-trained using a set of labelled data. In real-world, open-ended deployments, however, it is unrealistic to assume to have a pre-trained model for all possible environments. Therefore, agents need to dynamically learn/adapt/extend their perceptual abilities online, in an autonomous way, by exploring and interacting with the environment where they operate. This paper describes a way to do so, by exploiting symbolic planning. Specifically, we formalize the problem of automatically training a neural network to recognize object properties as a symbolic planning problem (using PDDL). We use planning techniques to produce a strategy for automating the training dataset creation and the learning process. Finally, we provide an experimental evaluation in both a simulated and a real environment, which shows that the proposed approach is able to successfully learn how to recognize new object properties.

Planning for Learning Object Properties

In this article, we show how a system for automated deduction can be given computational reflection, i.e., can affect its own computation mechanism, by using the very same machinery implementing logical deduction. This feature, which we call computational reflection via mechanized logical deduction, provides both theoretical and practical advantages. First, the theorem prover can inspect, extend, and modify its own underlying theorem-proving strategies automatically. Second, mechanized logical deduction can be used to reason about the ways these strategies can be extended and modified and to prove correctness statements. This opens up the possibility of building systems that are able to perform correct and safe, reflective self-extension and self-modification.

Computational reflection via mechanized logical deduction

A significant problem for integrating acting and planning is how to maintain consistency between the planner's descriptive action models, which abstractly describe what the actions do, and the actor's operational models, which tell how to perform the actions with rich control structures for closed-loop online decision-making. Operational models allow for dealing with a variety of contexts, and responding to unexpected outcomes and events in a dynamically changing environment. To circumvent the consistency problem, we use the actor's operational models both for acting and for planning. Our acting-and-planning algorithm, APE, uses hierarchical operational models inspired from those in the well-known PRS system. But unlike the reactive PRS algorithm, APE chooses its course of action using a planner that does Monte Carlo sampling over simulated executions of APE's operational models. Our experiments with this approach show significant benefits in the success rates of the acting system, in particular for domains with dead ends.

/pdf/using-operational-models-to-integrate-acting-and-planning-nft12w9arz.pdf

Paolo Traverso

Papers

From Propositional Deciders to First Order Deciders: a Structured Approach to the Decision Problem

Online Learning of Action Models for PDDL Planning

Planning for Learning Object Properties

Computational reflection via mechanized logical deduction

Using Operational Models to Integrate Acting and Planning