Options
Repairing neural networks for image classification problems using spectrum-based fault localization
Publikationstyp
Conference Paper
Date Issued
2024-06
Sprache
English
Author(s)
Rashedi, Daniel
Start Page
1
End Page
5
Citation
4th IEEE International Conference on Software Engineering and Artificial Intelligence, SEAI 2024
Contribution to Conference
Publisher DOI
Scopus ID
Publisher
IEEE
ISBN
979-8-3503-7435-3
979-8-3503-7434-6
Fault localization in neural networks develops methods to improve a model's accuracy, resilience, and performance. Improving a model's accuracy without extensive gradient descent training loops has not been a goal, yet. In this paper, we propose a novel model repair approach identifying suspicious neurons to be selected for modifications of greater impact upon a single iteration avoiding extensive training processes with the gradient descent method yielding potential overfitting. Our approach uses Spectrum-Based Fault Localization to guide the repair process. Spectrum-Based Fault Localization is an established method that assigns software code a suspiciousness value indicating a likelihood for fault presence depending on test suite outcomes and the associated code's execution. We employ hit-spectrum analysis to assign neurons a suspiciousness score. Our algorithm selects especially suspicious neurons for repair by reducing their activation values, while increasing the activation values of less suspicious neurons. Experiments conducted on convolutional layer and dense layer image classification models confirm that, overall, our approach leads to better mean accuracies. In 2 scenarios, the repair approach has led to small accuracy improvements by increasing the corresponding models' mean accuracies by 0.68 and 0.46 percentage points respectively. Remarkably, our repair approach has also led to significant accuracy improvements by increasing a specific model's accuracy by 15.73 percentage points. In conclusion, our method provides a promising alternative to additional training epochs running the risk of overfitting or depending on larger databases for the training process.
Subjects
hit-spectrum analysis
model pruning
suspiciousness score
suspicious neuron detection
DDC Class
004: Computer Sciences