Unmanned Surface Vehicle (USV) has a broad application prospect and autonomous path planning as its crucial technology has developed into a hot research direction in the field of USV research. This paper proposes an Improved Dueling Deep Double-Q Network Based on Prioritized Experience Replay (IPD3QN) to address the slow and unstable convergence of traditional Deep Q Network (DQN) algorithms in autonomous path planning of USV. Firstly, we use the deep double Q-Network to decouple the selection and calculation of the target Q value action to eliminate overestimation. The prioritized experience replay method is adopted to extract experience samples from the experience replay unit, increase the utilization rate of actual samples, and accelerate the training speed of the neural network. Then, the neural network is optimized by introducing a dueling network structure. Finally, the soft update method is used to improve the stability of the algorithm, and the dynamic $ϵ$-greedy method is used to find the optimal strategy. The experiments are first conducted in the Open AI Gym test platform to pre-validate the algorithm for two classical control problems: the Cart pole and Mountain Car problems. The impact of algorithm hyperparameters on the model performance is analyzed in detail. The algorithm is then validated in the Maze environment. The comparative analysis of simulation experiments shows that IPD3QN has a significant improvement in learning performance regarding convergence speed and convergence stability compared with DQN, D3QN, PD2QN, PDQN, PD3QN. Also, USV can plan the optimal path according to the actual navigation environment with the IPD3QN algorithm.

Image degradation technique has been the focus of current research in image super-resolution( SR). The High-order Deterioration Model(HDM) of Real-ESRGAN is more effective than the conventional bicubic kernel interpolation in simulating the degradation of real-world images. However, the images reconstructed by Real-ESRGAN suffer from two notable weaknesses. Firstly, the rebuilt image is overly smooth and suffers from substantial texture information loss, making it worse than classical models like SRGAN and ESRGAN. Moreover, despite the reconstructed images having a better visualization effect, they are entirely different from the original image, violating the principle of image reconstruction. In order to address these two issues, an improved model built on Real-ESRGAN (IRE) is presented in this work. We remove the first-order degradation modeling of HDM and keep its second-order degradation modeling to lessen the degree of visual deterioration. Then we utilized PatchGAN as the fundamental structure of the discriminator, added a channel attention mechanism to the dense block of the generator, and increased the texture detail in the reconstructed images. Finally, we replaced the L1 loss function with the SmoothL1 loss function to improve the convergence speed with better model performance. The ablation study shows that the average training time was decreased by about 28% when employing PatchGAN rather than U-net as the discriminator. In addition, the ablation study presents that the RankIQA measures assessed by IRE on Set5, Set14, and DIV2K100 were decreased by 15.20%, 20.46%, and 10.27%, respectively, and the PI measures considered by IRE+ on BSD100, Urban100 were decreased by 15.76%, 8.67%, respectively, compared to Real-ESRGAN. Also, the comparison experiment illustrates that the NIQE measures assessed by our suggested model on Set5, RealSR-Canon, and RealSR-Nikon were decreased by 12.31%, 3%, and 2.41%, respectively, compared to Real-ESRGAN.

The Ty1 retrotransposons present in the genome of Saccharomyces cerevisiae belong to the large class of mobile genetic elements that replicate via an RNA intermediary and constitute a significant portion of most eukaryotic genomes. The retromobility of Ty1 is regulated by numerous host factors, including several subunits of the Mediator transcriptional co-activator complex. In spite of its known function in the nucleus, previous studies have implicated Mediator in the regulation of post-translational steps in Ty1 retromobility. To resolve this paradox, we systematically examined the effects of deleting non-essential Mediator subunits on the frequency of Ty1 retromobility and levels of retromobility intermediates. Our findings reveal that loss of distinct Mediator subunits alters Ty1 retromobility positively or negatively over a >10,000-fold range by regulating the ratio of an internal transcript, Ty1i, to the genomic Ty1 transcript. Ty1i RNA encodes a dominant negative inhibitor of Ty1 retromobility that blocks virus-like particle maturation and cDNA synthesis. These results resolve the conundrum of Mediator exerting sweeping control of Ty1 retromobility with only minor effects on the levels of Ty1 genomic RNA and the capsid protein, Gag. Since the majority of characterized intrinsic and extrinsic regulators of Ty1 retromobility do not appear to effect genomic Ty1 RNA levels, Mediator could play a central role in integrating signals that influence Ty1i expression to modulate retromobility. Retrotransposons are mobile genetic elements that copy their RNA genomes into DNA and insert the DNA copies into the host genome. These elements contribute to genome instability, control of host gene expression and adaptation to changing environments. Retrotransposons depend on numerous host factors for their own propagation and control. The retrovirus-like retrotransposon, Ty1, in the yeast Saccharomyces cerevisiae has been an invaluable model for retrotransposon research, and hundreds of host factors that regulate Ty1 retrotransposition have been identified. Non-essential subunits of the Mediator transcriptional co-activator complex have been identified as one set of host factors implicated in Ty1 regulation. Here, we report a systematic investigation of the effects of loss of these non-essential subunits of Mediator on Ty1 retrotransposition. Our findings reveal a heretofore unknown mechanism by which Mediator influences the balance between transcription from two promoters in Ty1 to modulate expression of an autoinhibitory transcript known as Ty1i RNA. Our results provide new insights into host control of retrotransposon activity via promoter choice and elucidate a novel mechanism by which the Mediator co-activator governs this choice.

Performances of smartphones are profoundly affected by battery life. Maximizing the amount of usage of energy is essential to extend battery life. However, developers might concentrate more on the functionality of applications while ignoring the energy bugs that drain the battery during the development process. There are no quantitative approaches to detect these energy bugs introduced in this fast-paced development process. In this paper, we employ a system-call based approach to develop a power consumption model for Android devices. Data that measures the energy consumption of mobile devices under different testing scenarios with the number of triggered system calls are utilized in the model training process. A balanced recursive feature elimination with cross-validation approach is proposed to select and rank the importance of the different system calls. Seven machine learning models are trained over the selected features with cross-validation and hyper-parameter tuning technique, where linear regression with Lasso regularization outperforms all the other models. Then the model is evaluated on the data set that measures the energy consumption on different revision history of selected apps. Results show that the optimized Lasso model could detect energy bugs in the revision history of various applications. Optimization strategies are provided based on the selected features.

Fault diagnosis techniques, through the collection and analysis of fault information, effectively prevent frequent equipment failures. The widespread adoption of sensors facilitates data-driven diagnosis using extensive device data. However, these data are often unlabeled, and fault data are scarce, which limits the generalization capability of data-driven models. To overcome this, we utilize fault synthesis to generate ample training data. Despite the discrepancies between synthetic and real faults, Unsupervised Domain Adaptation (UDA) offers a viable solution. Existing research primarily focuses on global adaptation but inadequately addresses the internal structures of data, posing challenges in imbalanced fault scenarios. This paper introduces an innovative integration of domain adversarial training with structured prediction pseudo-labeling, employing Local Preserving Projection (LPP) to bridge the gap between synthetic and real fault data. Our proposed framework, validated using a synthetic dataset derived from the Case Western Reserve University (CWRU) bearing dataset, uses healthy samples as a foundation and incorporates expert knowledge to create a balanced fault dataset, ensuring its reliability. The framework achieves identification accuracies of 80.52% on balanced data and 83.94% on imbalanced data, surpassing the current state-of-the-art by 5.71%. The effectiveness of structured prediction pseudo-labels is further confirmed through ablation experiments.

Swin Transformers have been designed and used in various image super-resolution (SR) applications. One of the recent image restoration methods is RSTCANet, which combines Swin Transformer with Channel Attention. However, for some channels of images that may carry less useful information or noise, Channel Attention cannot automatically learn the insignificance of these channels. Instead, it tries to enhance their expression capability by adjusting the weights. It may lead to excessive focus on noise information while neglecting more essential features. In this paper, we propose a new image SR method, RSVTCANet, based on an extension of Swin2SR. Specifically, to effectively gather global information for the channel of images, we modify the Residual SwinV2 Transformer blocks in Swin2SR by introducing the coordinate attention for each two successive SwinV2 Transformer Layers (S2TL) and replacing Multi-head Self-Attention (MSA) with Efficient Multi-head Self-Attention version 2 (EMSAv2) to employ the resulting residual SwinV2 Transformer coordinate attention blocks (RSVTCABs) for feature extraction. Additionally, to improve the generalization of RSVTCANet during training, we apply an optimized RandAugment for data augmentation on the training dataset. Extensive experimental results show that RSVTCANet outperforms the recent image SR method regarding visual quality and measures such as PSNR and SSIM.