Interpretable human-label-free deep learning for real-bogus classification with uncertainty quantification

Time-domain surveys generate large numbers of transient candidates, making real-bogus classification a critical step in automated discovery pipelines. Obtaining reliable human-labeled training sets is costly, while community-provided labels can be noisy and survey-dependent. We aim to develop a real-bogus classification framework for point-like astrophysical transient detections against bogus survey detections, that can be trained without human-labeled data using physically motivated injected transients and bogus-dominated survey data, that remains robust under strong class contamination, and that provides well-calibrated and reliable uncertainty quantification (UQ) suitable for downstream decision-making. We combine simulated transient injections with a heavily contaminated survey class and train a dual-network model using an asymmetric co-teaching strategy for classes with different label-noise levels. We evaluate performance on a labeled benchmark subset and analyze the learned representation with latent-space visualization tools. For UQ, we compare standard approaches (Monte Carlo dropout and deep ensembles) and propose a low-cost hybrid strategy that exploits the dual-network setting to improve calibration. We further extend the evaluation to the light-curve domain to assess recovery of light-curve classes. The method achieves strong real-bogus performance on the labeled subset and remains stable under severe class contamination. It recovers transient light-curve classes with high fidelity, while single-source identification is fundamentally limited by intrinsic ambiguity in light-curve-derived labels. Our hybrid UQ approach achieves competitive calibration relative to substantially more expensive ensemble baselines. Latent-space analyses indicate that uncertainty aligns with the decision boundary and reveal that the model can identify subclasses within the bogus population. Our results show that injection-driven, weakly supervised training can enable scalable and consistent real-bogus classification without human-labeled training data while providing calibrated uncertainties. The methodology is well suited for transfer to forthcoming surveys by rerunning the injection-based training pipeline.