Haiyi Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Haiyi Liu has a total of 16 co-authors. Here are the top co-authors by number of shared publications:
- Shaoying Liu
East China Normal University
3 shared publications Ai Liu
Nanjing University of Aeronautics and Astronautics
1 shared publication- Ai Liu
College of Computer Science and Technology, Nanjing University of Aeronautics and Astronautics
1 shared publication - Dingbang Fang
Graduate School of Advanced Science and Engineering, Hiroshima University Hiroshima 739-8527, Japan
1 shared publication - Dingbang Fang
Fujian Normal University
1 shared publication Guangquan Xu
School of Cybersecurity, College of Intelligence and Computing, Tianjin University Tianjin 300072, China
1 shared publication
Latest Publications
- Open Access
- 19 January 2025
- Published by MDPI AG in Electronics
- Vol. 14 (2) , 382
Mutation-Based Approach to Supporting Human–Machine Pair Inspection
Human–machine pair inspection refers to a technique that supports programmers and machines working together as a “pair” in source code inspection tasks. The machine provides guidance, while the programmer performs the inspection based on this guidance. Although programmers are often best suited to inspect their own code due to familiarity, overconfidence may lead them to overlook important details. This study introduces a novel mutation-based human–machine pair inspection method, which is designed to direct the programmer’s attention to specific code components by applying targeted mutations. We assess the effectiveness of code inspections by analyzing the programmer’s corrections of these mutations. Our approach involves defining mutation operators for each keyword in the program based on historical defects, developing mutation rules based on program keywords and a strategy for automatically generating mutants, and designing a code comparison strategy to quantitatively evaluate code inspection quality. Through a controlled experiment, we demonstrate the effectiveness of mutation-based human–machine pair inspection in aiding programmers during the inspection process.
- 19 January 2025
- Published by MDPI AG in Electronics
- Vol. 14 (2) , 382
Open AccessMutation-Based Approach to Supporting Human–Machine Pair Inspection
Human–machine pair inspection refers to a technique that supports programmers and machines working together as a “pair” in source code inspection tasks. The machine provides guidance, while the programmer performs the inspection based on this guidance. Although programmers are often best suited to inspect their own code due to familiarity, overconfidence may lead them to overlook important details. This study introduces a novel mutation-based human–machine pair inspection method, which is designed to direct the programmer’s attention to specific code components by applying targeted mutations. We assess the effectiveness of code inspections by analyzing the programmer’s corrections of these mutations. Our approach involves defining mutation operators for each keyword in the program based on historical defects, developing mutation rules based on program keywords and a strategy for automatically generating mutants, and designing a code comparison strategy to quantitatively evaluate code inspection quality. Through a controlled experiment, we demonstrate the effectiveness of mutation-based human–machine pair inspection in aiding programmers during the inspection process.
- 17 October 2023
- Published by World Scientific Pub Co Pte Ltd in International Journal of Software Engineering and Knowledge Engineering
- Vol. 34 (02) , 273-300
NNTBFV: Simplifying and Verifying Neural Networks Using Testing-Based Formal Verification
Neural networks are extensively employed in safety-critical systems. However, these critical systems incorporating neural networks continue to pose risks due to the presence of adversarial examples. Although the security of neural networks can be enhanced by verification, verifying neural networks is an NP-hard problem, making the application of verification algorithms to large-scale neural networks a challenging task. For this reason, we propose NNTBFV, a framework that utilizes the principles of Testing-Based Formal Verification (TBFV) to simplify neural networks and verify the simplified networks. Unlike conventional neural network pruning techniques, this approach is based on specifications, with the goal of deriving approximate execution paths under given preconditions. To mitigate the potential issue of unverifiable conditions due to overly broad preconditions, we also propose a precondition partition method. Empirical evidence shows that as the range of preconditions narrows, the size of the execution paths also reduces accordingly. The execution path generated by NNTBFV is still a neural network, so it can be verified by verification tools. In response to the results from the verification tool, we provide a theoretical method for analysis. We evaluate the effectiveness of NNTBFV on the ACAS Xu model project, choosing Verification-based and Random-based neural network simplification algorithms as the baselines for NNTBFV. Experiment results show that NNTBFV can effectively approximate the baseline in terms of simplification capability, and it surpasses the efficiency of the random-based method.
- 17 October 2023
- Published by World Scientific Pub Co Pte Ltd in International Journal of Software Engineering and Knowledge Engineering
- Vol. 34 (02) , 273-300
NNTBFV: Simplifying and Verifying Neural Networks Using Testing-Based Formal Verification
Neural networks are extensively employed in safety-critical systems. However, these critical systems incorporating neural networks continue to pose risks due to the presence of adversarial examples. Although the security of neural networks can be enhanced by verification, verifying neural networks is an NP-hard problem, making the application of verification algorithms to large-scale neural networks a challenging task. For this reason, we propose NNTBFV, a framework that utilizes the principles of Testing-Based Formal Verification (TBFV) to simplify neural networks and verify the simplified networks. Unlike conventional neural network pruning techniques, this approach is based on specifications, with the goal of deriving approximate execution paths under given preconditions. To mitigate the potential issue of unverifiable conditions due to overly broad preconditions, we also propose a precondition partition method. Empirical evidence shows that as the range of preconditions narrows, the size of the execution paths also reduces accordingly. The execution path generated by NNTBFV is still a neural network, so it can be verified by verification tools. In response to the results from the verification tool, we provide a theoretical method for analysis. We evaluate the effectiveness of NNTBFV on the ACAS Xu model project, choosing Verification-based and Random-based neural network simplification algorithms as the baselines for NNTBFV. Experiment results show that NNTBFV can effectively approximate the baseline in terms of simplification capability, and it surpasses the efficiency of the random-based method.
- Open Access
- 9 December 2022
- Published by Frontiers Media SA in Frontiers in Neuroscience
- Vol. 16, 1096737
How far is brain-inspired artificial intelligence away from brain?
Fueled by the development of neuroscience and artificial intelligence (AI), recent advances in the brain-inspired AI have manifested a tipping-point in the collaboration of the two fields. AI began with the inspiration of neuroscience, but has evolved to achieve a remarkable performance with little dependence upon neuroscience. However, in a recent collaboration, research into neurobiological explainability of AI models found that these highly accurate models may resemble the neurobiological representation of the same computational processes in the brain, although these models have been developed in the absence of such neuroscientific references. In this perspective, we review the cooperation and separation between neuroscience and AI, and emphasize on the current advance, that is, a new cooperation, the neurobiological explainability of AI. Under the intertwined development of the two fields, we propose a practical framework to evaluate the brain-likeness of AI models, paving the way for their further improvements.
- 9 December 2022
- Published by Frontiers Media SA in Frontiers in Neuroscience
- Vol. 16, 1096737
Open AccessHow far is brain-inspired artificial intelligence away from brain?
Fueled by the development of neuroscience and artificial intelligence (AI), recent advances in the brain-inspired AI have manifested a tipping-point in the collaboration of the two fields. AI began with the inspiration of neuroscience, but has evolved to achieve a remarkable performance with little dependence upon neuroscience. However, in a recent collaboration, research into neurobiological explainability of AI models found that these highly accurate models may resemble the neurobiological representation of the same computational processes in the brain, although these models have been developed in the absence of such neuroscientific references. In this perspective, we review the cooperation and separation between neuroscience and AI, and emphasize on the current advance, that is, a new cooperation, the neurobiological explainability of AI. Under the intertwined development of the two fields, we propose a practical framework to evaluate the brain-likeness of AI models, paving the way for their further improvements.