PAFL: Probabilistic Automaton-based Fault Localization for Recurrent Neural Networks

Ishimoto, Y.; Kondo, Masanari; Ubayashi, Naoyasu; Kamei, Yasutaka

doi:10.1016/j.infsof.2022.107117

Cited by 9 publications

(4 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Assume DeepCNN localizes the root causes (DU), which is the number of units in the Dense layer. In this case, the algorithm will rewrite the source code and define the hyperparameters while building the models as follows: Dense(units=hp.Int('units', [16,32,64,128,256,512,1024]), activation='relu'). Then, the algorithm starts to explore different combinations of hyperparameters by building N models with different hyperparameter combinations and tries to find the best model during the hyperparameter tuning process (line-4).…”

Section: Automated Cnn Model Repairmentioning

confidence: 99%

DeepCNN: A Dual Approach to Fault Localization and Repair in Convolutional Neural Networks

Wardat,

Al-Alaj

2024

IEEE Access

View full text Add to dashboard Cite

Deep learning models, particularly Convolutional Neural Networks (CNNs), play a pivotal role in intelligent software. However, like any software application, CNN-based applications are susceptible to bugs. Bug-fix patterns in CNN differ from traditional techniques, primarily due to their inherent blackbox nature. Moreover, current methods, although tailored for generic DNN structures, are time consuming, require specialized expertise, and are not directly applicable to the unique structure and requirements of CNNs. To address these issues, we propose DeepCNN, an innovative automated tool to identify the root causes of CNN faults and autonomously fix prevalent training faults that impact the performance and efficiency of CNN programs. DeepCNN, a data-driven tool, employs a transformer encoder model to abstract token-level CNN code, enabling it to effectively detect and fix bugs in real-world CNN models. Beyond mere identification, DeepCNN offers automated solutions for repairing CNN hyperparameter misconfigurations using an optimization solver. Additionally, by analyzing data dependencies and employing a search algorithm, it determines the most optimal hyperparameter values for the problematic models, ultimately generating patch fixes. Through rigorous evaluation on 36 diverse buggy models, DeepCNN outperformed existing methods in fault localization and repair, showing a robustness in identifying potential problems with a 90% detection rate. Among these problematic models, it manages to repair 90% of them -resulting in a 30% average accuracy boost.

show abstract

Section: Automated Cnn Model Repairmentioning

confidence: 99%

DeepCNN: A Dual Approach to Fault Localization and Repair in Convolutional Neural Networks

Wardat,

Al-Alaj

2024

IEEE Access

View full text Add to dashboard Cite

show abstract

“…In the automata theory, an automaton does not refer to a concrete machine but rather a dynamic mathematical model of discrete systems that can transform and process information [27,28]. Probabilistic finite automata (PFA) is a branch of automata that is a structured representation of discrete mathematical models.…”

Section: Pfamentioning

confidence: 99%

Particle Swarm Optimization-Based Model Abstraction and Explanation Generation for a Recurrent Neural Network

Liu,

Wang,

2024

Algorithms

View full text Add to dashboard Cite

In text classifier models, the complexity of recurrent neural networks (RNNs) is very high because of the vast state space and uncertainty of transitions, which makes the RNN classifier’s explainability insufficient. It is almost impossible to explain the large-scale RNN directly. A feasible method is to generalize the rules undermining it, that is, model abstraction. To deal with the low efficiency and excessive information loss in existing model abstraction for RNNs, this work proposes a PSO (Particle Swarm Optimization)-based model abstraction and explanation generation method for RNNs. Firstly, the k-means clustering is applied to preliminarily partition the RNN decision process state. Secondly, a frequency prefix tree is constructed based on the traces, and a PSO algorithm is designed to implement state merging to address the problem of vast state space. Then, a PFA (probabilistic finite automata) is constructed to explain the RNN structure with preserving the origin RNN information as much as possible. Finally, the quantitative keywords are labeled as an explanation for classification results, which are automatically generated with the abstract model PFA. We demonstrate the feasibility and effectiveness of the proposed method in some cases.

show abstract

“…Software testing is a critical step in ensuring software quality and reliability and can effectively reduce the enormous economic losses and personnel safety risks caused by software faults [1]. In the software development process, traditional manual debugging requires huge human resources and time costs, making automated testing particularly important [2,3]. In this process, software fault localization is a key step.…”

Section: Introductionmentioning

confidence: 99%

“…Sci. 2023, 13, 12344 2 of 21 fault localization techniques [2,8,9], slicing-based program localization techniques [10,11], information retrieval-based program localization techniques [12], and machine learningbased fault localization techniques [3,13]. These methods have undoubtedly contributed to the research of program fault localization, but there is still a large amount of information and features hidden within the program itself and during the program running process that have not been explored by current research.…”

Section: Introductionmentioning

confidence: 99%

Fault Localization Using TrustRank Algorithm

Fan,

Wu,

Zhang

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

Spectrum-based fault localization (SBFL), a widely recognized technique in automated fault localization, has limited effectiveness due to its disregard for the internal information of the program under test suites. To overcome this limitation, a novel TrustRank-based fault localization (TRFL) technique is introduced. TRFL enhances traditional SBFL by incorporating internal data dependencies of the program under the test suite, thereby providing a more comprehensive analysis. It constructs a node-weighted program execution network and employs the TrustRank algorithm to analyze network centrality and re-rank program entities based on their suspiciousness. Furthermore, a bidirectional TrustRank algorithm (Bi-TRFL) is extended that takes into account the influence relationship between network nodes for more accurate fault localization. When applied to large-scale datasets with real faults, such as Defects4J, TRFL, and Bi-TRFL, it significantly outperforms traditional SBFL methods in fault localization. They demonstrate up to 40% and 13% improvement in Top-1 and Top-5 rankings, respectively, proving their robustness and efficiency with minimal sensitivity to related parameters.

show abstract

PAFL: Probabilistic Automaton-based Fault Localization for Recurrent Neural Networks

Cited by 9 publications

References 37 publications

DeepCNN: A Dual Approach to Fault Localization and Repair in Convolutional Neural Networks

DeepCNN: A Dual Approach to Fault Localization and Repair in Convolutional Neural Networks

Particle Swarm Optimization-Based Model Abstraction and Explanation Generation for a Recurrent Neural Network

Fault Localization Using TrustRank Algorithm

Contact Info

Product

Resources

About