Flight safety remains a central concern in civil aviation. Recently, increasing attention has been given to leveraging high-dimensional temporal flight data, typically collected by Quick Access Recorders (QAR), to predict safety events. However, most existing studies focus on individual events, overlooking the latent correlations among multiple events. For instance, during landing, a maneuver that reduces the risk of one event may inadvertently increase the risk of another. Predicting multiple events introduces three main challenges: 1)handling parameters recorded at inconsistent sampling rates, 2) learning task-specific parameter importance for interpretability, and 3) modeling complex temporal dependencies for accuracy. To address these, we propose MultiSafe, a deep multi-task learning model for predicting and interpreting multiple flight safety events. First, to process parameters with heterogeneous frequencies, we introduce a Multi-Scale Shared Encoder with adaptive convolutional kernels to unify representations across tasks. Next, a gating-based Parameter Selector learns task-specific parameter importance, enabling interpretable predictions. Finally, a Temporal Decoder with fine-grained attention captures intricate temporal dependencies among parameters. Experiments on a dataset of 37,518 A320 flight records demonstrate that MultiSafe outperforms state-of-the-art baselines in prediction accuracy, while its interpretability offers actionable insights to assist pilots in enhancing flight safety.

$\ell_1$-penalized quantile regression ($\ell_1$-QR) is a useful tool for modeling the relationship between input and output variables when detecting heterogeneous effects in the high-dimensional setting. Hypothesis tests can then be formulated based on the debiased $\ell_1$-QR estimator that reduces the bias induced by Lasso penalty. However, the non-smoothness of the quantile loss brings great challenges to the computation, especially when the data dimension is high. Recently, the convolution-type smoothed quantile regression (SQR) model has been proposed to overcome such shortcoming, and people developed theory of estimation and variable selection therein. In this work, we combine the debiased method with SQR model and come up with the debiased $\ell_1$-SQR estimator, based on which we then establish confidence intervals and hypothesis testing in the high-dimensional setup. Theoretically, we provide the non-asymptotic Bahadur representation for our proposed estimator and also the Berry-Esseen bound, which implies the empirical coverage rates for the studentized confidence intervals. Furthermore, we build up the theory of hypothesis testing on both a single variable and a group of variables. Finally, we exhibit extensive numerical experiments on both simulated and real data to demonstrate the good performance of our method.

Recent techniques have achieved remarkable improvements depended on mining subtle yet distinctive features for fine-grained visual classification (FGVC). While prior works directly combine discriminative features extracted from different parts, we argue that the potential interactions between different parts and their abilities to category predictions should be taken into consideration, which enables significant parts to contribute more to the decision of the sub-category. To this end, we present a Cross-Part Convolutional Neural Network (CP-CNN) in a weakly supervised manner to explore cross-learning among multi-regional features. Specifically, the context transformer is implemented to encourage joint feature learning across different parts under the guidance of a navigator. The part with the highest confidence is regarded as a navigator to deliver distinguishing characteristics to the others with lower confidence while the complementary information is retained. To locate discriminative but subtle parts precisely, a part proposal generator (PPG) is designed with the feature enhancement blocks, through which complex scale variations caused by the viewpoint diversity can be effectively alleviated. Extensive experiments on three benchmark datasets demonstrate that our proposed method consistently outperforms existing state-of-the-art methods.