End-to-end face quality assessment based on deep learning can directly predict the overall quantitative score of face quality, thus helping to control the risk of face recognition system. Thanks to the development of automatic quality pseudo-label generation, most recent methods can use large-scale face datasets to learn the quality model. However, existing methods use regression models to fit the pseudo-labels, which lack attention to samples that are easy to be misidentified, and require large models for training. The paper treats the quality assessment as a classification problem, focusing on difficult samples near the classification boundary. Specifically, pairwise binary quality pseudo-label is generated based on the face similarity score without additional manual annotation. An identification quality loss is used to decouple the pairwise network training. In addition, a lightweight quality network is trained by performing knowledge distillation on the quality prediction branch of the face recognition network. Experiments show that the proposed quality network achieves state-of-the-art results with only 0.45 M parameters and 77 M FLOPs.

When dealing with UAV path planning problems, evolutionary algorithms demonstrate strong flexibility and global search capabilities. However, as the number of UAVs increases, the scale of the path planning problem grows exponentially, leading to a significant rise in computational complexity. The Cooperative Co-Evolutionary Algorithm (CCEA) effectively addresses this issue through its divide-and-conquer strategy. Nonetheless, the CCEA needs to find a balance between computational efficiency and algorithmic performance while also resolving convergence difficulties arising from the increased number of decision variables. Moreover, the complex interrelationships between the decision variables of each UAV add to the challenge of selecting appropriate decision variables. To tackle this problem, we propose a novel collaborative algorithm called CCEA-ADVS. This algorithm reduces the difficulty of the problem by decomposing high-dimensional variables into sub-variables for collaborative optimization. To improve the efficiency of decision variable selection in the collaborative algorithm and to accelerate the convergence speed, an adaptive decision variable selection strategy is introduced. This strategy selects decision variables according to the order of solving single-UAV constraints and multi-UAV constraints, reducing the cost of the optimization objective. Furthermore, to improve computational efficiency, a two-stage evolutionary optimization process from coarse to fine is adopted. Specifically, the Adaptive Differential Evolution with Optional External Archive algorithm (JADE) is first used to optimize the waypoints of the UAVs to generate a basic path, and then, the Dubins algorithm is combined to optimize the trajectory, yielding the final flight path. The experimental results show that in four different scenarios involving 40 UAVs, the CCEA-ADVS algorithm significantly outperforms the Grey Wolf Optimizer (GWO), Particle Swarm Optimization (PSO), Artificial Bee Colony (ABC), and JADE algorithms in terms of path performance, running time, computational efficiency, and convergence speed. In addition, in large-scale experiments involving 500 UAVs, the algorithm also demonstrates good adaptability, stability, and scalability.

Boost-phase trajectory inference is one of the major objectives of the space-borne missile early-warning system. Traditional methods can be classified as either profile-based or profile-free methods. The profile-based methods are accurate but inadaptable to the types of missiles, whereas the profile-free methods are adaptable but inaccurate. To integrate the strengths of the profile-based and profile-free methods, a multimodel trajectory inference approach is proposed. First, a general net acceleration model (GNAM) containing only type-free prior information is constructed by the method of sieves. Then, a new kind of net acceleration profile is proposed by incorporating type-dependent prior information into the GNAM. After that, the multimodel approach is proposed following the Bayesian framework. Simulations indicate that the approach is accurate in estimation and capable for type identification.

Estimating the boost-phase trajectory of a ballistic missile using line of sight measurements from space-borne passive sensors is an important issue in missile defense. A well-known difficulty of this issue is the poor-observability of the target motion. A profile-based maximum penalised likelihood estimator is presented, which is expected to work in poor-observability scenarios. Firstly, a more adaptable boost-phase profile is proposed by introducing unknown parameters. Then, the estimator is given based on the Bayesian paradigm. After that, a special penalty for box constraint is constructed based on a mixed distribution. Numerical results for some typical scenarios and sensitivity with respect to a priori information are reported to show that the proposed estimator is promising.

In the last 20 years, a lot of achievements have been made in the study of posterior contraction rates of nonparametric Bayesian methods, and plenty of them involve sieve priors, but mainly for specific models or sieves. We provide a posterior contraction theorem for general parametric sieve priors. The theorem has weaker and simpler conditions compared with the existing results, and indicates that the sieve prior is rate adaptive. We apply the general theorem to density estimations and nonparametric regression with jumps. We also provided a reversible jump MCMC (Markov Chain Monte Carlo) algorithm for the sieve prior.