Search

In application, training data and test data collected via indoor positioning algorithms usually do not come from the same ideal conditions. Changes in various environmental conditions and signal drift can cause different probability distributions between the data sets. Existing positioning algorithms cannot guarantee stable accuracy when facing these issues, resulting in dramatic reduction and the infeasibility of the positioning accuracy of indoor location algorithms. Considering these restrictions, domain adaptation technology in transfer learning has proven to be a promising solution in past research in terms of solving the inconsistent probability distribution problems. However, most localization algorithms based on transfer learning do not perform well because they only learn a shallow representation feature, which can only slightly reduce the domain discrepancy. Based on the deep network and its strong feature extraction ability, it can learn more transferable features for domain adaptation and achieve better domain adaptation effects. A Deep Joint Mean Distribution Adaptation Network (DJMDAN) is proposed to align the global domain and relevant subdomain distributions of activations in multiple domain-specific layers across domains to achieve domain adaptation. The test results demonstrate that the performance of the proposed method outperforms the comparison algorithm in indoor positioning applications. [ABSTRACT FROM AUTHOR]

Image super-resolution reconstruction is a research hotspot in the field of computer vision. Traditional image super-resolution reconstruction methods based on deep learning mostly up-sample low-resolution images ignoring categories and instances, which will cause some problems such as unrealistic texture in the reconstructed images or sawtooth phenomenon on the edge of instance. In this manuscript, we propose an image super-resolution reconstruction method based on instance spatial feature modulation and feedback mechanism. First, the prior knowledge of instance spatial features is introduced in the reconstruction process. Instance spatial features of low-resolution images are extracted to modulate super-resolution reconstruction features. Then, based on the feedback mechanism, the modulated low-resolution image features are iteratively optimized for the reconstruction results, so that the model can finally learn instance-level reconstruction ability. Experiments on COCO-2017 show that, compared with traditional deep learning-based image super-resolution reconstruction methods, the proposed method can obtain better image reconstruction results, and the reconstructed images have more realistic instance textures. [ABSTRACT FROM AUTHOR]

In recent years, even though Stochastic Gradient Descent (SGD) and its variants are well-known for training neural networks, it suffers from limitations such as the lack of theoretical guarantees, vanishing gradients, and excessive sensitivity to input. To overcome these drawbacks, alternating minimization methods have attracted fast-increasing attention recently. As an emerging and open domain, however, several new challenges need to be addressed, including 1) Convergence properties are sensitive to penalty parameters, and 2) Slow theoretical convergence rate. We, therefore, propose a novel monotonous Deep Learning Alternating Minimization (mDLAM) algorithm to deal with these two challenges. Our innovative inequality-constrained formulation infinitely approximates the original problem with non-convex equality constraints, enabling our convergence proof of the proposed mDLAM algorithm regardless of the choice of hyperparameters. Our mDLAM algorithm is shown to achieve a fast linear convergence by the Nesterov acceleration technique. Extensive experiments on multiple benchmark datasets demonstrate the convergence, effectiveness, and efficiency of the proposed mDLAM algorithm. [ABSTRACT FROM AUTHOR]

Contour error compensation of the computer numerical control (CNC) machine tool is a vital technology that can improve machining accuracy and quality. To achieve this goal, the tracking error of a feeding axis, which is a dominant issue incurring the contour error, should be firstly modeled and then a proper compensation strategy should be determined. However, building the precise tracking error prediction model is a challenging task because of the nonlinear issues like backlash and friction involved in the feeding axis; besides, the optimal compensation parameter is also difficult to determine because it is sensitive to the machining tool path. In this paper, a set of novel approaches for contour error prediction and compensation is presented based on the technologies of deep learning and reinforcement learning. By utilizing the internal data of the CNC system, the tracking error of the feeding axis is modeled as a Nonlinear Auto-Regressive Long-Short-Term Memory (NAR-LSTM) network, considering all the nonlinear issues of the feeding axis. Given the contour error as calculated based on the predicted tracking error of each feeding axis, a compensation strategy is presented with its parameters identified efficiently by a Time-Series Deep Q-Network (TS-DQN) as designed in our work. To validate the feasibility and advantage of the proposed approaches, extensive experiments are conducted, testifying that our approaches can predict the tracking error and contour error with very good precision (better than about 99% and 90% respectively), and the contour error compensated based on the predicted results and our compensation strategy is significantly reduced (about 60~85% reduction) with the machining quality improved drastically (machining error reduced about 50%). [ABSTRACT FROM AUTHOR]

Selecting a set of initial users from a social network in order to maximize the envisaged number of influenced users is known as influence maximization (IM). Researchers have achieved significant advancements in the theoretical design and performance gain of several classical approaches, but these advances are almost reaching their pinnacle. Learning-based IM approaches have emerged recently with a higher generalization to unknown graphs than conventional methods. The development of learning-based IM methods is still constrained by a number of fundamental hardships, including (1) solving the objective function efficiently, (2) struggling to characterize the diverse underlying diffusion patterns, and (3) adapting the solution to different node-centrality-constrained IM variants. To address the aforementioned issues, we design a novel framework DeepIM for generatively characterizing the latent representation of seed sets, as well as learning the diversified information diffusion pattern in a data-driven and end-to-end way. Subsequently, we design a novel objective function to infer optimal seed sets under flexible node-centrality-based budget constraints. Extensive analyses are conducted over both synthetic and real-world datasets to demonstrate the overall performance of DeepIM. [ABSTRACT FROM AUTHOR]