DOI: 10.3897/jucs.207838

Authors: Shiguang Guo, ZhiHao Liu

Abstract: Fuzzing has emerged as a vital technique for discovering vulnerabilities in software systems. American Fuzzy Lop (AFL) is a widely used fuzzer known for its effectiveness and efficiency. However, optimizing fuzzing strategies to enhance coverage and bug discovery remains a significant challenge. This paper introduces an innovative approach by integrating an Adaptive Reward Mechanism within the Proximal Policy Optimization (PPO) framework to improve AFL fuzzing performance. Our method dynamically adjusts rewards based on the fuzzing feedback, guiding the learning agent to prioritize paths with higher potential for uncovering vulnerabilities. Experimental evaluations demonstrate a clear trend of improvement in both code coverage and unique crashes found as model iterations increase. The Adaptive PPO method consistently outperforms the traditional AFL and standard PPO techniques across all iterations. By the 100th iteration, Adaptive PPO achieves 90% code coverage and discovers 25 unique crashes, compared to AFL's 60% coverage and 11 crashes, and PPO's 78% coverage and 17 crashes. This data highlights the effectiveness of the adaptive reward mechanism in enhancing exploration and vulnerability detection capabilities, supporting the assertion that Adaptive PPO is a superior approach for fuzz testing. The incremental improvements observed with each iteration underscore the method's potential for sustained performance gains over time, advancing the state-of-the-art in software testing and vulnerability analysis. Additionally, performance overhead analysis shows the method’s practical feasibility." to emphasize overhead validation.

HTML

PDF

DOI: 10.3897/jucs.106023

Authors: Alberto Jiménez-Macías, Pedro J. Muñoz-Merino, Margarita Ortiz-Rojas, Mario Muñoz-Organero, Carlos Delgado Kloos

Abstract: Educational content has become a key element for improving the quality and effectiveness of teaching. Many studies have been conducted on user and knowledge modeling using machine-learning algorithms in smart-learning environments. However, few studies have focused on content modeling to estimate content indicators based on student interaction. This study presents a systematic literature review of content modeling using machine learning algorithms in smart learning environments. Two databases were used: Scopus and Web of Science (WoS), with studies conducted until August 2023. In addition, a manual search was performed at conferences and in relevant journals in the area. The results showed that assessment was the most used content in the papers, with difficulty and discrimination as the most common indicators. Item Response Theory (IRT) is the most commonly used technique; however, some studies have used different traditional learning algorithms such as Random Forest, Neural Networks, and Regression. Other indicators, such as time, grade, and number of attempts, were also estimated. Owing to the few studies on content modeling using machine learning algorithms based on interactions, this study presents new lines of research based on the results obtained in the literature review.

HTML

PDF

DOI: 10.3217/jucs-023-06-0529

Authors: Marat Faizrahmanov, Iskander Kalimullin, Antonio Montalban, Vadim Puzarenko

Abstract: Studying the ∑-reducibility of families introduced by [Kalimullin and Puzarenko 2009] we show that for every set there is a family of sets which is the ∑-least countable family whose ∑-jump is ∑-equivalent to . This fact will be generalized for the class of n-families (families of families of ... of sets).

HTML

XML

PDF

DOI: 10.3217/jucs-022-07-0943

Authors: Marat Faizrahmanov, Iskander Kalimullin

Abstract: In this paper we introduce a hierarchy of families which can be derived from the integers using countable collections. This hierarchy coincides with the von Neumann hierarchy of hereditary countable sets in the ZFC-theory with urelements from ℕ. The families from the hierarchy can be coded into countable algebraic structures preserving their algorithmic properties. We prove that there is no maximal level of the hierarchy and that the collection of non-lowα degrees for every computable ordinal ff is the enumeration spectrum of a family from the hierarchy. In particular, we show that the collection of non-lowα degrees for every computable limit ordinal α is a degree spectrum of some algebraic structure.

HTML

XML

PDF

DOI: 10.3217/jucs-021-12-1563

Authors: Cristian S. Calude, Rūsiņš Freivalds, Sanjay Jain, Frank Stephan

Abstract: A set L is (m, n)-computable iff there is a mechanism which on input of n different words produces n conjectures whether these words are in L, respectively, such that at least m of these conjectures are right. Prior studies dealt with (m, n)- computable sets in the contexts of recursion theory, complexity theory and the theory of finite automata. The present work aims to do this with respect to computations by deterministic pushdown automata (using one common stack while processing all input words in parallel). We prove the existence of a deterministic context-free language L which is recognised by an (1, 1)-DPDA but fails to be recognised by any (m, n)-DPDA, where n ≥ 2 and m ≥ n/2+1. This answers a question posed by Eli Shamir at LATA 2013. Furthermore, it is shown that there is a language L such that, for all m, n with m ≤ n/2, L can be recognised by an (m, n)-DPDA but, for all m, n with 1 ≤ m ≤ n, L cannot be recognised by (m, n)-DFA.

HTML

XML

PDF

DOI: 10.3217/jucs-021-11-1425

Authors: Hsing-Wen Wang, Yen-Chun Wu, Tse-Ping Dong

Abstract: Social networking websites have become increasingly popular, and have also become the main media not only to connect lives socially, but also to affect brand image and consumers' purchase intention. The purpose of this paper is to incorporate the Facebook fan page and e-journal provide over the Internet (cloud e-journal) with the uses and gratification theory to test the impact on brand image and purchase intention through the use of cloud learning. We used cloud learning material from the Ivy League in Taiwan in our case study. This paper also applied structural equation modeling to analyze the data collected from members of the Ivy League Facebook fan page and the Cloud users’ e-journal. The results of this study demonstrated that for the members of the Ivy League Facebook fan page, purchase intention was positively and significantly influenced, regardless of any use intention factors, based on the uses and gratification theory. In addition, using the Facebook fan page and Cloud e-journal would also positively and significantly affect the brand image for Internet users. Moreover, with the Ivy League fan page and the Cloud e-journal's improved brand image, there is an increase in intention to buy the journals and relevant services. This paper also demonstrated that six features of the Cloud e-journal did have a moderating effect on the purchase intention. Our results provide suggestions to those who attempt to build cloud learning solutions for customers, and are also helpful to those who wish to apply the Facebook fan page to customer relationship marketing platforms.

HTML

XML

PDF

DOI: 10.3217/jucs-018-14-1987

Authors: Jörg Keller, Christoph Kessler, Rikard Hultén

Abstract: Limited bandwidth to off-chip main memory tends to be a performance bottleneck in chip multiprocessors, and this will become even more problematic with an increasing number of cores. Especially for streaming computations where the ratio between computational work and memory transfer is low, transforming the program into more memory-efficient code is an important program optimization. On-chip pipelining reorganizes the computation so that partial results of subtasks are forwarded immediately between the cores over the high-bandwidth internal network, in order to reduce the volume of main memory accesses, and thereby improves the throughput for memory-intensive computations. At the same time, throughput is also constrained by the limited amount of on-chip memory available for buffering forwarded data. By optimizing the mapping of tasks to cores, balancing a trade-off between load balancing, buffer memory consumption, and communication load on the on-chip network, a larger buffer size can be applied, resulting in less DMA communication and scheduling overhead. In this article, we consider parallel mergesort as a representative memory-intensive application in detail, and focus on the global merging phase, which is dominating the overall sorting time for larger data sets. We work out the technical issues of applying the on-chip pipelining technique, and present several algorithms for optimized mapping of merge trees to the multiprocessor cores. We also demonstrate how some of these algorithms can be used for mapping of other streaming task graphs. We describe an implementation of pipelined parallel mergesort for the Cell Broadband Engine, which serves as an exemplary target. We evaluate experimentally the influence of buffer sizes and mapping optimizations, and show that optimized on-chip pipelining indeed speeds up, for realistic problem sizes, merging times by up to 70% on QS20 and 143% on PS3 compared to the merge phase of CellSort, which was by now the fastest merge sort implementation on Cell.

HTML

XML

PDF

DOI: 10.3217/jucs-017-06-0874

Authors: Lynette van Zijl, Jaco Geldenhuys

Abstract: We investigate ambiguity for symmetric difference nondeterministic finite automata. We show the existence of unambiguous, finitely ambiguous, polynomially ambiguous and exponentially ambiguous symmetric difference nondeterministic finite automata. We show that, for each of these classes, there is a family of n-state nondeterministic finite automata such that the smallest equivalent deterministic finite automata have O(2n) states.

HTML

XML

PDF

DOI: 10.3217/jucs-016-18-2563

Authors: Christine Gaßner

Abstract: We consider the uniform BSS model of computation where the machines can perform additions, multiplications, and tests of the form x ≥ 0. The oracle machines can also check whether a tuple of real numbers belongs to a given oracle set or not. We present oracle sets containing positive integers and pairs of numbers, respectively, such that the classes P and DNP relative to these oracles are not equal. The first set is constructed by diagonalization techniques and the second one is derived from the Knapsack Problem.

HTML

XML

PDF

DOI: 10.3217/jucs-016-15-1929

Authors: Ichiro Satoh

Abstract: This paper presents an agent-based system for building and operating agent-basedcontext-aware services in public spaces, including museums. The system provides users with agents and detects the locations of users and deploys location-aware user-assistant agents at com-puters near the their current locations by using active RFID-tags. When a visitor moves between exhibits in a museum, this dynamically deploys his/her agent at the computers close to the ex-hibits by using mobile agent technology. It annotates the exhibits in his/her personalized form and navigate him/her user to the next exhibits along his/her routes. It also introduces user move-ment as a natural approach to interacting between users and agents. To demonstrate the utility and effectiveness of the system, we constructed location/user-aware visitor-guide services andexperimented them for two weeks in a public museum.

HTML

XML

PDF

DOI: 10.3217/jucs-015-06-1186

Authors: Christine Gaßner

Abstract: We consider the uniform model of computation over any structure with two constants. For several structures, we construct oracles which imply that the relativized versions of P and NP are equal or are not equal. We construct universal oracles which imply the equality of the relativized versions of P and NP and we show that we lose the possibility to define these oracles recursively if we try to compress their elements to tuples of fixed length. Moreover we give new oracles for the BSS model in order to separate the classes P and NP relative to these oracles.

HTML

XML

PDF

DOI: 10.3217/jucs-014-15-2471

Authors: Sadiq Sait, Khawar Khan, Mustafa Ali

Abstract: This paper discusses the parallelization of Stochastic Evolution (StocE) metaheuristic, for a distributed parallel environment. VLSI cell placement is used as an optimization problem. A comprehensive set of parallelization approaches are tested and an effective strategy is identified in terms of two underlying factors: workload division and the effect of parallelization on metaheuristic's search intelligence. The strategies are compared with parallelization of another similar evolutionary metaheuristic called Simulated Evolution (SimE). The role of the two mentioned underlying factors is discussed in parallelization of StocE.

HTML

XML

PDF

DOI: 10.3217/jucs-014-06-0956

Authors: Volker Bosserhoff

Abstract: We define and compare several probabilistic notions of computability for mappings from represented spaces (that are equipped with a measure or outer measure) into computable metric spaces. We thereby generalize definitions by [Ko 1991] and Parker (see [Parker 2003, Parker 2005, Parker 2006]), and furthermore introduce the new notion of computability in the mean. Some results employ a notion of computable measure that originates in definitions by [Weihrauch 1999] and [Schröder 2007]. In the spirit of the well-known Representation Theorem (see [Weihrauch 2000]), we establish dependencies between the probabilistic computability notions and classical properties of mappings. We furthermore present various results on the computability of vector-valued integration, composition of mappings, and images of measures. Finally, we discuss certain measurability issues arising in connection with our definitions.

HTML

XML

PDF

DOI: 10.3217/jucs-012-11-1521

Authors: David Escrig, Carlos Rodríguez

Abstract: Bisimulation can be defined in a simple way using coinductive methods, and has rather pleasant properties. Ready similarity was proposed by Meyer et al. as a way to weakening the bisimulation equivalence thus getting a semantics defined in a similar way, but supported for more reasonable (weaker) observational properties. Global bisimulations were introduced by Frutos et al. in order to study different variants of non-determinism getting, in particular, a semantics under which the internal choice operator becomes associative. Global bisimulations are defined as plain bisimulations but allowing the use of new moves, called global transitions, that can change the processes not only locally in its head, but anywhere. Now we are continuing the study of global bisimulation but focusing on the way different semantics can be characterised as global bisimulation semantics. In particular, we have studied ready similarity, on the one hand because it was proposed as the strongest reasonable semantics weaker than bisimulation; on the other hand, because ready similarity was not directly defined as an equivalence relation but as the nucleus of an order relation, and this open the question whether it is also possible to define it as a symmetric bisimulation-like semantics. We have got a simple and elegant characterisation of ready similarity as a global bisimulation semantics, that provides a direct symmetric characterisation of it as an equivalence relation, without using any order as intermediate concept. Besides, we have found that it is not necessary to start from a simulation based semantics to get an equivalent global bisimulation. What has proved to be very useful is the axiomatic characterisation of the semantics. Following these ideas we have got also global bisimulation for several semantics, including refusals and traces. That provides a general framework that allows to relate both intensional and extensional semantics.

HTML

XML

PDF

DOI: 10.3217/jucs-012-06-0746

Authors: Harald Hempel

Abstract: In this paper we give an introduction to the connection between complexity theory and the study of randomized algorithms. In particular, we will define and study probabilistic complexity classes, survey the basic results, and show how they relate to the notion of randomized algorithms.

HTML

XML

PDF

DOI: 10.3217/jucs-012-06-0725

Authors: Tobias Riege, Jörg Rothe

Abstract: NP-complete problems cannot have efficient algorithms unless P = NP. Due to their importance in practice, however, it is useful to improve the known exponential-time algorithms for NP-complete problems. We survey some of the recent results on such improved exponential-time algorithms for the NP-complete problems satisfiability, graph colorability, and the domatic number problem. The deterministic time bounds are compared with the corresponding time bounds of randomized algorithms, which often run faster but only at the cost of having a certain error probability.

HTML

XML

PDF

DOI: 10.3217/jucs-012-06-0654

Authors: Johannes Blõmer

Abstract: We give a brief introduction to probabilistic encryptions. This serves as an example how randomness plays a pivotal role in cryptographic systems that satisfy advanced security concepts.

HTML

XML

PDF

DOI: 10.3217/jucs-012-05-0551

Authors: Tobias Riege, Jörg Rothe

Abstract: This paper surveys some of the work that was inspired by Wagner's general technique to prove completeness in the levels of the boolean hierarchy over NP and some related results. In particular, we show that it is DP-complete to decide whether or not a given graph can be colored with exactly four colors, where DP is the second level of the boolean hierarchy. This result solves a question raised by Wagner in 1987, and its proof uses a clever reduction due to Guruswami and Khanna. Another result covered is due to Cai and Meyer: The graph minimal uncolorability problem is also DP-complete. Finally, similar results on various versions of the exact domatic number problem are discussed.

HTML

XML

PDF

DOI: 10.3217/jucs-011-09-1494

Authors: José Bravo, Ramón Hervás, Gabriel Chavira

Abstract: In recent years, there have been many efforts at research towards obtaining the simple and natural use of computers, with interfaces closer to the user. New visions such as that of the Ubiquitous Computing paradigm emerge. In Ubiquitous Computing the computer is distributed in a series of devices with reduced functionality, spread over the user's environment and communicating wirelessly. With these, context-aware applications are obtained. In this paper we present an approach to the classroom context by identification process using RFID technology, as an implicit input to the system. The main goal is to acquire natural interaction, because the only requirement for the user (teacher or student) is to carry a device (smart label), identifying and obtaining context services. Some of these services and the mechanisms that make them available are described here, together with a scenario of their use in the classroom.

HTML

XML

PDF

DOI: 10.3217/jucs-008-04-0423

Authors: Mirko Viroli, Andrea Omicini

Abstract: Observation is a fundamental interaction pattern in today's computer-based systems. Adopting observation as the main modelling criterion, computer-based systems can be represented as composed by three class of entities: observers, observables (or sources), and coordinators, that is, the entities managing the observer/source interaction. Also, agents and agent societies are fundamental abstractions in modelling today's complex systems. When exploiting observation in the context of agent-based systems, the most natural interpretation for agents is to see them as either observers or coordinators. However, their situatedness and autonomy, their peculiar perception and representation of the environment, and their typical ability to infer new knowledge - in short, their individual viewpoint over the world -, make agents suitable for an interpretation as observable sources. Accordingly, this paper discusses the implications of using observation to model agent systems, and focuses on the interpretation of agents as observables. A formal framework is developed where multiagent systems are modelled as the composition of agents interacting by observing each other and by mutually affecting their observable behaviour.

HTML

XML

PDF

DOI: 10.3217/jucs-008-01-0002

Authors: Egon Börger

Abstract: The research belonging to the Abstract State Machines approach to system design and analysis is surveyed and documented in an annotated ASM bibliography. The survey covers the period from 1984, when the idea for the concept of ASMs (under the name dynamic or evolving algebras or structures) appears for the first time in a foundational context, to the year 2001 where a mathematically well-founded, practical system development method based upon the notion of ASMs is in place and ready to be industrially deployed. Some lessons for the future of ASMs are drawn.

HTML

XML

PDF

DOI: 10.3217/jucs-007-11-0917

Authors: Yuri Gurevich, Nikolai Tillmann

Abstract: The partial update problem for parallel abstract state machines has manifested itself in the cases of counters, sets and maps. We propose a solution of the problem that lends itself to an efficient implementation and covers the three cases mentioned above. There are other cases of the problem that require a more general framework.

HTML

XML

PDF

DOI: 10.3217/jucs-006-05-0490

Authors: Horia Georgescu

Abstract: This paper presents a new model for the specification of communicating X-machine systems (CXMS). In previous papers, systems of X-machines have been implemented in two ways: using an unique X-machine which simulates the concurrent behaviour of several processes [1], or using several X-machines which communicate through asynchronous channels [2]. This article introduces an X-machine system for which the communication between components is done through synchronous channels. The model supposes that each X-machine has a local memory, an input and an output tape. The X-machines act simultaneously. The states of each component of the system are partitioned into processing and communicating states. Passing messages between the X-machines involves only communicating states. It is shown that, taking advantage of the behaviour of X-machines, communication using channels may be implemented, thus providing a synchronous message passing.

HTML

XML

PDF

DOI: 10.3217/jucs-005-09-0494

Authors: Tudor Balanescu, Anthony Cowling, Horia Georgescu, Marian Gheorghe, Mike Holcombe, Cristina Vertan

Abstract: A version of the communicating stream X-machine model is proposed, which gives a precise representation of the operation of transferring data from one X-machine to another. For this model it is shown that systems of communicating X-machines have the same computational power as single stream X-machines. This enable existing methods for deriving test strategies for stream X-machines to be extended to communicating stream X-machines.

HTML

XML

PDF

DOI: 10.3217/jucs-003-04-0233

Authors: Yuri Gurevich, Marc Spielmann

Abstract: According to the ASM thesis, any algorithm is essentially a Gurevich abstract state machine. The only objection to this thesis, at least in its sequential version, has been that ASMs do not capture recursion properly. To this end, we suggest recursive ASMs. 1.) Partially supported by NSF grant CCR 95-04375 and ONR grant N00014-94-1-1182. 2.) Visiting scholar at the University of Michigan, partially supported by DAAD and The University of Michigan.

HTML

XML

PDF

DOI: 10.3217/jucs-001-01-0023

Authors: Ville Leppänen

Abstract: We show how to implement an -processor EREW PRAM workoptimally on a 2-dimensional n-sided mesh of trees, consisting of n processors, n memory modules, and nodes. Similarly, we prove that an -processor EREW PRAM can be implemented work-optimally on a 3-dimensional n-sided mesh of trees. By the work-optimality of implementations we mean that the expected routing time of PRAM memory requests is per simulated PRAM processor with high probability. Experiments show that on relatively small and the cost per simulated PRAM processor is 1.5-2.5 in the 2-dimensional case, and 2-3 in the 3-dimensional case. If at each step at most 1/3'th of the PRAM processors make a reference to the shared memory, then the simulation cost is approximately 1. We also compare our work-optimal simulations to those proposed for coated meshes.

HTML

XML

PDF