GAssert A Fully Automated Tool to Improve Assertion Oracles

GASSERT is the first tool to automatically improve assertion oracles . Assertions are prone to both false positives and false negatives . Given a Javamethod containing an assertion oracle to improve, Gassert returns an improved assertion with fewer false positives, fewer false negatives than the initial assertion .…

Robustness Evaluation of Stacked Generative Adversarial Networks using Metamorphic Testing

Synthesising photo-realistic images from natural language is one of thechallenging problems in computer vision . Stacked GenerativeAdversarial Network (StackGAN-v2) has proven capable of generating highresolution images that reflect the details specified in the input textdescriptions . The proposed MetamorphicTesting technique can be applied to other text-to-image synthesis models tonot only verify the robustness but also help researchers understand andinterpret the results made by the machine learning models.…

Extreme mutation testing in practice An industrial case study

Mutation testing is used to evaluate the effectiveness of test suites . Extreme mutation testing identifies methods where their functionality can be entirely removed, and the test suite would not notice it . These methods are called pseudo-tested . In this paper, we compare execution and analysis times for traditional and extrememutation testing and discuss what they mean in practice .…

Natural Hoare Logic Towards formal verification of programs from logical forms of natural language specifications

Formal verification provides strong guarantees of correctness of software, especially in safety or security critical systems . Hoarelogic is a widely used formalism for rigorous verification of software againstspecifications in the form of pre-condition/post-condition assertions . This paper proposes a formal framework for Hoare logic-based formalverification of imperative programs using logical forms generated fromcompositional semantic parsing of natural language assertions .…

Sublinear Domination and Core Periphery Networks

In this paper we devise a generative random network model with core-peripheryproperties whose core nodes act as sublinear dominators . We show that instances generated by this model exhibit power lawdegree distributions, and incorporates small-world phenomena . We also fit ourmodel in a variety of real-world networks .…

Weisfeiler and Lehman Go Topological Message Passing Simplicial Networks

Message Passing Simplicial Networks (MPSNs) perform message passing on simplicial complexes (SCs) Weisfeiler-Lehman (SWL)colouring procedure for distinguishing non-isomorphic SCs . We show that MPSNs are strictly more powerful than the WL test and not less powerfulthan the 3-WL test . We empirically supportour theoretical claims by showing that MpsNs can distinguish challenging regular graphs for which GNNs fail .…

Deep Graph Structure Learning for Robust Representations A Survey

Graph Neural Networks (GNNs) are widely used for analyzing graph-structured data . Most GNN methods are highly sensitive to the quality of graph structures . However, the pervasiveness of noise in graphs necessitates learningrobust representations for real-world problems . Many studies have been proposed around the central concept of GraphStructure Learning (GSL), which aims to jointly learn an optimized graphstructure and corresponding representations .…

Deep Recurrent Encoder A scalable end to end network to model brain signals

Deep Recurrent Encoding (DRE) architecture reliably predicts MEGresponses to words with a three-fold improvement over classic linear methods . We successfully test this approach on a large cohort ofmagnetoencephalography (MEG) recordings acquired during a one-hour reading task . The quantitative improvement of the present deep learning approachpaves the way to better understand the nonlinear dynamics of brain activity from large datasets .…

Self play Learning Strategies for Resource Assignment in Open RAN Networks

Open Radio Access Network (ORAN) is being developed with an aim to democratise access and lower the cost of future mobile data networks . In ORAN, network functionality is dis-aggregated into remote units(RUs) and distributed units (DUs) A deepreinforcement learning-based self-play approach is proposed to achieve efficient RU-DU resource management, with AlphaGo Zero inspired neuralMonte-Carlo Tree Search (MCTS) The self-played learning strategy is said to achieve intelligent resource assignment for different network conditions .…

Towards Fully Intelligent Transportation through Infrastructure Vehicle Cooperative Autonomous Driving Challenges and Opportunities

The infrastructure-vehicle cooperative autonomous driving approach depends on cooperation between intelligent roads and intelligent vehicles . Thisapproach is not only safer but also more economical compared to the traditionalon vehicle-only autonomous driving . This approach is safer and more economical than traditional autonomous driving approaches .…

Online adaptive algorithm for Constraint Energy Minimizing Generalized Multiscale Discontinuous Galerkin Method

In this research, we propose an online basis enrichment strategy within the framework of a recently developed constraint energy minimizing generalizedmultiscale discontinuous Galerkin method (CEM-GMsDGM) Combining the techniqueof oversampling, one makes use of the information of the current residuals toadaptively construct basis functions in the online stage .…

Surrogate assisted cooperative signal optimization for large scale traffic networks

Reasonable setting of traffic signals can be very helpful in alleviatingcongestion in urban traffic networks . SCSO first decomposes it into a set of tractable sub-networks, and then achieves signal setting by cooperatively optimizing these sub-netships . The decomposition operation significantlynarrows the search space of the whole traffic network, and thesurrogate-assisted optimizer greatly lowers the computational burden by reducing the number of expensive traffic simulations .…

Functional Extensionality for Refinement Types

Refinement type checkers are a powerful way to reason about functionalprograms . Without functional extensionality, proofs must relate functions that are fully applied . When working with first-order data,fully applied proofs lead to noisome duplication when using higher-orderfunctions . We define a propositional equality in a library we call PEq .…

An Axiomatic Approach to Detect Information Leaks in Concurrent Programs

Realizing flow security in a concurrent environment is extremely challenging,primarily due to non-deterministic nature of execution . The difficulty isfurther exacerbated from a security angle if sequential threads disclose control locations through publicly observable statements like print, sleep,delay, etc. Such observations lead to internal and external timing attacks .…

Energy and Cost Efficient Resource Allocation for Blockchain Enabled NFV

Network function virtualization (NFV) is a promising technology to make 5Gnetworks flexible and agile . In NFV, users’ service request can be viewed as a service function chain (SFC)consisting of several virtual network functions . Resource allocation in NFV is done through a centralizedauthority called NFV Orchestrator (NFVO) This centralized authority suffers from some drawbacks such as single point of failure and security .…

How to Identify Boundary Conditions with Contrasty Metric

The boundary conditions (BCs) have shown great potential in requirements engineering because a BC captures the particular combination of circumstances,i.e., divergence, in which the goals of the requirement cannot be satisfied as a whole . Existing researches have attempted to automatically identify lots ofBCs.…

Graph Information Vanishing Phenomenon inImplicit Graph Neural Networks

The basic idea of implicit GNNs is to introduce graphinformation with special properties followed by Learnable TransformationStructures (LTS) which encode the importance of neighbor nodes via adata-driven way . LTS maps different graph information into highly similar results . The relationship between graph information and LTS should be rethought to ensure that graph information is used in noderepresentation .…

On entanglement assistance to a noiseless classical channel

For a classical channel, neither the Shannon capacity nor the sum ofconditional probabilities corresponding to the cases of successful transmission can be increased by the use of a non-signaling resource . Yet, perhaps somewhat counterintuitively, entanglement assistance can help and actually elevate thechances of success even in a one-way communicational task that is to becompleted by a single-shot use of the noiseless classical channel .…

Estimating the Expected Influence Capacities of Nodes in Complex Networks under the Susceptible Infectious Recovered SIR Model

In recent years, epidemic modeling in complex networks has found many applications, including modeling of information or gossip spread in onlinesocial networks . In this context, it isvery important to identify super-spreaders that maximize or minimizepropagation . We developed an algorithm that approximates the expected influence of nodes under the popular SIR model .…

Reconstructed spatial receptive field structures by reverse correlation technique explains the visual feature selectivity of units in deep convolutional neural networks

An important issue in dealing with Deep Convolutional Neural Networks (DCNN) is the ‘black box problem’, which represents the unknowns about internal information representation and processing . In the middle layers (convolutional layers in block3 and block4), AWCanalysis successfully reconstructed the receptive field that predicted the visual feature selectivity of the unit .…

Root cause prediction based on bug reports

Knowing the root cause of a bug can help developers in the debugging process . We mined 54755 closed bug reports from issue trackers of 103 GitHub projects and applied a set of heuristicsto create a benchmark . A subset was manuallyclassified into three groups (semantic, memory, and concurrency) based on thebugs’ root causes .…

Multi view Audio and Music Classification

The proposed multi-view network consists of four subnetworks, each handling one input type . The learned embedding are then concatenated to form the multi view embedding for classification similar to a simpleconcatenation network . A novel method is proposed to keep track of the learning behavior on the classification branches and adapt their weights toproportionally blend their gradients for network training .…