TransitNet: A Compact Attention-Augmented Deep Learning Framework for Low-SNR Transit Blind Searches

2026-06-17

AI Summary

This study addresses the challenge of detecting mid- to long-period Earth-like exoplanet transit signals in low signal-to-noise ratio (SNR) regimes, where conventional methods such as TLS and BLS suffer from low detection rates. To overcome this limitation, the authors propose a lightweight attention-augmented deep learning framework—the first to incorporate attention mechanisms into transit signal detection. Leveraging a unified simulation pipeline, rigorous evaluation via ROC and PR metrics, and optimized inference design, the model achieves 95.2% detection accuracy in the SNR range of 6–8 with a compact size of approximately 1.5 MB. In injection-recovery experiments, it attains a 93.0% recovery rate for Earth-analog planets, operates 12–25× faster than CPU-based TLS, and successfully recovers all 34 known Kepler planets, substantially outperforming traditional approaches.

Abstract

Motivated by the observational incompleteness of intermediate-to-long-period Earth-size planets, we present TransitNet, a compact attention-augmented deep-learning framework for low-SNR transit blind searches. To enable realistic method development and objective threshold calibration under blind-search conditions, we develop a unified dataset construction, benchmarking, and threshold-selection framework. On recovery benchmarks constructed from unseen Kepler targets, TransitNet attains 95.2 percent accuracy in the challenging SNR range of 6 to 8 and outperforms both TLS and BLS, achieving ROC-AUC and PR-AP values of 0.974 and 0.982, respectively. In an injected Earth-size and sub-Earth-size transit recovery experiment, TransitNet achieves a recovery rate of 93.0 percent, substantially exceeding those of TLS (63.1 percent) and BLS (60.0 percent). In addition to detection, TransitNet provides attention-based estimates of transit windows and midpoints. On an independent evaluation set, 97.4 percent of injected transits are fully covered by the estimated transit window. Applied to real Kepler observations, the model successfully recovers all 34 selected confirmed Kepler planets, with a mean absolute transit midpoint error of 1.24 hours. The model combines a compact footprint of about 1.5 MB with high inference efficiency, yielding speed-ups of about 12 to 25 times relative to CPU-TLS and about 4 to 5 times relative to CPU-BLS. These results demonstrate that TransitNet provides an accurate, scalable, and computationally efficient framework for low-SNR transit blind searches in the tested regime and motivate its extension to longer-period Earth-size planet searches.

Problem

• transit blind search
• low-SNR
• Earth-size planets
• exoplanet detection
• observational incompleteness

Innovation

• attention mechanism
• low-SNR transit detection
• deep learning for exoplanets
• blind search framework
• computational efficiency