Reducing false positives in strong lens detection through effective augmentation and ensemble learning
This work addresses the challenge of improving accuracy in gravitational lens detection for astronomy missions like Euclid, representing an incremental advance through enhanced data handling and model techniques.
This research tackled the problem of reducing false positives in detecting strong gravitational lenses using CNNs, achieving a false positive rate of 10^-4 and identifying over 88% of genuine lenses, which is an 11-fold reduction in false positives with only a 2.3% decrease in true positives.
This research studies the impact of high-quality training datasets on the performance of Convolutional Neural Networks (CNNs) in detecting strong gravitational lenses. We stress the importance of data diversity and representativeness, demonstrating how variations in sample populations influence CNN performance. In addition to the quality of training data, our results highlight the effectiveness of various techniques, such as data augmentation and ensemble learning, in reducing false positives while maintaining model completeness at an acceptable level. This enhances the robustness of gravitational lens detection models and advancing capabilities in this field. Our experiments, employing variations of DenseNet and EfficientNet, achieved a best false positive rate (FP rate) of $10^{-4}$, while successfully identifying over 88 per cent of genuine gravitational lenses in the test dataset. This represents an 11-fold reduction in the FP rate compared to the original training dataset. Notably, this substantial enhancement in the FP rate is accompanied by only a 2.3 per cent decrease in the number of true positive samples. Validated on the KiDS dataset, our findings offer insights applicable to ongoing missions, like Euclid.