In this paper, we further advance a line of work aimed to formally model via epistemic logic (aspects of) the group dynamics of cooperative agents. In fact, we have previously proposed and here extend a particular log...
详细信息
Scenario-based software requirements specifications, due to limitations of natural language and scenarios, lack precision and abstraction. Formal methods address this problem, but are rarely used. A Unified Object-Ori...
详细信息
C and C++ are widely-used, mature programming languages. they have been extensively used in development of projects such as Linux, Windows, YouTube, Adobe, Firefox, and Google Chrome. Due to poor memory safety, C and ...
详细信息
ISBN:
(纸本)9798350353020;9798400705779
C and C++ are widely-used, mature programming languages. they have been extensively used in development of projects such as Linux, Windows, YouTube, Adobe, Firefox, and Google Chrome. Due to poor memory safety, C and C++ programs are vulnerable to security attacks as are programs in languages that depend on C/C++ library code. As per the Common Weakness Enumeration (CWE), out-of-bounds (OOB) write in C/C++ programs topped the list of the 25 most dangerous software weakness in 2021 and 2022. Fixing OOB write at the source code level still requires human experts. this is a tedious task that may result in erroneous programs. In this paper we propose a technique to create a data set of corresponding flawed and correct programs that can be used to perform supervised training of deep-learning models to automate the process of detecting and patching OOB writes. the proposed technique has two elements: collecting a set of C/C++ programs from online sources (correct programs) and injecting OOB write errors into them, thus yielding a set of corresponding flawed programs. In this paper we focus on four main flaws associated with OOB writes: faulty access, faulty declaration, faulty guard in loops, and faulty usage of memory-write APIs. We have found that popular fault localization tools can not localize complicated bugs in our buffer overflow sample set (BOSS). In addition, the current state-of-the-art machine learning security flaw repair tool could not repair any of the bugs in a randomly selected set of BOSS samples and, in some cases, generated out-of-bound writes as suggested patches. these results lead us to conclude that the bugs injected by our tool are significant and our dataset is useful for training neural program repair models. We also propose two data-augmentation techniques to overcome problems associated with limited-size corpora.
Due to the recent rapid introduction of AI technologies into society, we face new risks related to AI. therefore, it is very important to let AI be compliant with legal and ethical norms to reduce such risks. In this ...
详细信息
the proceedings contain 45 papers. the topics discussed include: show me your attach request and I'll tell you who you are: practical fingerprinting attacks in 4G and 5G mobile networks;a scary peek into the futur...
ISBN:
(纸本)9781665421416
the proceedings contain 45 papers. the topics discussed include: show me your attach request and I'll tell you who you are: practical fingerprinting attacks in 4G and 5G mobile networks;a scary peek into the future: advanced persistent threats in emerging computing environments;reliability models and analysis for triple-model with triple-input machine learning systems;security orchestration, automation, and response engine for deployment of behavioral honeypots;a co-evolutionary algorithm-based malware adversarial sample generation method;securing password authentication for web-based applications;graph neural network-based android malware classification with jumping knowledge;multi-task learning model based on multiple characteristics and multiple interests for CTR prediction;device-to-device task offloading in a stochastic invalid-device scenario with social awareness;and a novel approach for providing client-verifiable and efficient access to private smart contracts.
We present profile and experience of a hybrid seminar on fundamental issues of software engineering, theory and experimental programming ru-STEP (= russian seminar on Software Engineering, theory and Experimental Prog...
详细信息
In my PhD thesis, the concept of Cognitive agents with extended Learning capabilities for Autonomous Mobility on Demand (AMoD) scenarios is investigated. Specifically, the focus is set on the Ride-hailing concept with...
详细信息
the demand for formal verification tools for neural networks has increased as neural networks have been deployed in a growing number of safety-critical applications. Matrices are a data structure essential to formalis...
详细信息
ISBN:
(纸本)9783031212215;9783031212222
the demand for formal verification tools for neural networks has increased as neural networks have been deployed in a growing number of safety-critical applications. Matrices are a data structure essential to formalising neural networks. Functional programming languages encourage diverse approaches to matrix definitions. this feature has already been successfully exploited in different applications. the question we ask is whether, and how, these ideas can be applied in neural network verification. A functional programming language Imandra combines the syntax of a functional programming language and the power of an automated theorem prover. Using these two key features of Imandra, we explore how different implementations of matrices can influence automation of neural network verification.
Logic has been proved useful to model various aspects of the reasoning process of agents and multi-agentsystems (MAS). In this paper, we report about the last advances over a line of work aimed to explore social aspe...
详细信息
ISBN:
(数字)9783030974572
ISBN:
(纸本)9783030974572;9783030974565
Logic has been proved useful to model various aspects of the reasoning process of agents and multi-agentsystems (MAS). In this paper, we report about the last advances over a line of work aimed to explore social aspects of such systems. the objective is to formally model (aspects of) the group dynamics of cooperative agents. We have proposed and here extend a particular logical framework (the Logic of "Inferable" L-DINF), where a group of cooperative agents can jointly perform actions. I.e., at least one agent of the group can perform the action, either withthe approval of the group or on behalf of the group. We have been able to take into consideration actions' cost and the preferences that each agent can have for what concerns performing each action. Our focus here is on: (i) explainability, i.e., the syntax of our logic is especially devised to make it possible to transpose a proof into a natural language explanation, in the perspective of trustworthy Artificial Intelligence;(ii) the capability to construct and execute joint plans within a group of agents;(iii) the formalization of aspects of the theory of Mind, which is an important social-cognitive skill involving the ability to attribute mental states, including emotions, desires, beliefs, and knowledge to oneself and to others, and to reason about the practical consequences of such mental states;such capability is very relevant when agents have to interact with humans, and in particular in robotic applications;(iv) connection between theory and practice, so as to make our logic actually usable by a system's designers.
In this paper, we present EPMC, an extendible probabilistic model checker. EPMC has a small kernel, and is designed modularly. It supports discrete probabilistic models such as Markov chains and Markov decision proces...
详细信息
ISBN:
(数字)9783030945831
ISBN:
(纸本)9783030945824;9783030945831
In this paper, we present EPMC, an extendible probabilistic model checker. EPMC has a small kernel, and is designed modularly. It supports discrete probabilistic models such as Markov chains and Markov decision processes. Like PRISM, it supports properties specified in PCTL*. Two central advantages of EPMC are its modularity and extendibility. We demonstrate these features by extending EPMC to EPMC-PETL, a model checker for probabilistic epistemic properties on multi-agentsystems. EPMC-petl takes advantage of EPMC to provide two model checking algorithms for multi-agentsystems with respect to probabilistic epistemic logic: an exact algorithm based on SMT techniques and an approximated one based on UCT. multi-agentsystems and epistemic properties are given in an extension of the modelling language of PRISM, making it easy to model this kind of scenarios.
暂无评论