Intra coding plays a critical role in video compression, yet conventional directional intra prediction mainly relies on boundary pixels and fails to capture complex textures and long-range spatial correlations. Template Matching Prediction (TMP) alleviates this issue by searching similar templates in reconstructed regions without motion-vector signaling; however, its MSE/MAE-based matching and simple averaging fusion are sensitive to noise and often unreliable for complex content. This paper proposes E2ETMPN, an end-to-end template matching prediction network for intra coding. E2ETMPN jointly models template matching and multi-candidate fusion under a unified loss function. It consists of two components: (1) a CNN-based multiscale matching sub-network that searches candidate predictions from reconstructed regions at different spatial scales; and (2) a Transformer-based fusion sub-network that adaptively fuses multiple candidates with template information using attention mechanisms to generate the final prediction. Experimental results demonstrate that integrating E2ETMPN into the reference encoder achieves consistent BD-rate reductions over conventional TMP and standard intra prediction methods on standard test sequences. Notably, E2ETMPN yields larger gains on screen content and scenes with rich non-local repetitive structures, validating the effectiveness of end-to-end learning for template matching prediction in intra coding.

Video-Text Retrieval (VTR) aims to align video content with natural language descriptions and is a fundamental task in multi-modal understanding. Most existing methods model videos as uniformly sampled RGB frames, which overlooks rich temporal cues, especially motion dynamics encoded in videos. We propose Compressed Video Stream Learning for Video-Text Retrieval (CVSVTR), a framework that directly exploits information from compressed video streams to enhance retrieval performance without fully decoding videos. Specifically, CVSVTR decodes only the I-frames of each GOP and extracts appearance features using a CLIP-based encoder. Meanwhile, motion vectors and residuals are parsed from the compressed bitstream and processed by a lightweight P-frame Feature Generation (PFG) module to construct motion-aware representations for P-frames. A spatial-channel attention mechanism is further introduced to adaptively fuse appearance features with compressed-domain motion cues, compensating for temporal information missed by uniform frame sampling. Extensive experiments on the MSR-VTT and MSVD benchmarks demonstrate that CVSVTR consistently outperforms existing video-text retrieval methods across multiple evaluation metrics, validating the effectiveness of leveraging compressed video streams for efficient and accurate temporal modeling in VTR.

This study develops an advanced self-triggered predefined-time safety learning control algorithm for switched multi-agent systems with full-state mask. To strengthen encryption while reducing the impact to system performance, an improved settling time privacy preservation mechanism based on the full-state mask function is designed, which encrypts the true information of the system and enhances the privacy of information delivery. Unlike traditional learning control schemes, a novel actor-critic weight update law is designed to guarantee that the system energy cost is minimized resulting in predefined time optimization. Besides, an improved self-triggered condition with a compensation term is developed to overcome the complex challenges posed by full-state privacy preservation mechanism. It not only eliminates the necessity to continually monitor the triggered state of the system but also saves communication resources. Finally, the validity of the designed control scheme can be proven by a simulation experiment.

This article proposes a prescribed-time fuzzy optimal control approach for flexible-joint (FJ) robot systems utilizing the reinforcement learning (RL) strategy. The uniqueness of this method lies in its ability to ensure optimal tracking performance for n-link flexible joint robots within the prescribed-time frame, while the actor and critic fuzzy logic system effectively approximate the optimal cost and evaluates system performance. First, the optimal controllers with the auxiliary compensation term are constructed by utilizing the online approximation of the modified performance index function and RL actor-critic structure. The designed controller can deal with unknown structure impacts and avoid model identification. Besides, in designing the prescribed-time scale function, the introduced constant term not only prevents singularity but also allows flexible setting of constraint regions. The proposed scheme is theoretically verified to satisfy the Bellman optimality principle and ensure the tracking error converges to the desired zone within the prescribed time. Finally, the practicability of the designed control scheme is further demonstrated by the 2-link FJ robot simulation example.

This study presents an advanced adaptive fuzzy optimal bridge-hole constraint control method for large-scale interconnected systems under quantized input. To address the conflict in constraint ranges caused by the combined effect of both results in the bridge-hole and performance constraints, a new prescribed time function with parameter requirements is proposed, which bridges the balance between them and keeps the tracking error within a desired zone in a prescribed time. Meanwhile, output constraint is realized by building a new bridge-hole constraint function, which ensures the time interval for the constraint behavior to occur by the flexible setting of the switching time. Unlike traditional optimal control schemes, the designed optimal controller is further quantized by a hysteresis quantizer, which minimizes energy cost and saves bandwidth. Besides, a reinforcement learning (RL) scheme based on an actor–critic-identifier fuzzy logic system (FLS) structure is designed; its overall control idea is to optimize the entire backstepping control system by using all virtual and actual backstepping control as the optimal solution of their respective subsystems. Finally, the effectiveness of the proposed scheme is confirmed by simulation experiments.

Owing to its simplicity, the traditional k - means (Lloyd heuristic) clustering method plays a vital role in a variety of machine-learning applications. Disappointingly, the Lloyd heuristic is prone to local minima. In this article, we propose k - mRSR, which converts the sum-of-squared error (SSE) (Lloyd) into a combinatorial optimization problem and incorporates a relaxed trace maximization term and an improved spectral rotation term. The main advantage of k - mRSR is that it only needs to solve the membership matrix instead of computing the cluster centers in each iteration. Furthermore, we present a nonredundant coordinate descent method that brings the discrete solution infinitely close to the scaled partition matrix. Two novel findings from the experiments are that k - mRSR can further decrease (increase) the objective function values of the k - means obtained by Lloyd (CD), while Lloyd (CD) cannot decrease (increase) the objective function obtained by k - mRSR. In addition, the results of extensive experiments on 15 datasets indicate that k - mRSR outperforms both Lloyd and CD in terms of the objective function value and outperforms other state-of-the-art methods in terms of clustering performance.