Semi-supervised semantic segmentation has witnessed remarkable advancements in recent years. However, existing algorithms are based on convolutional neural networks and directly applying them to Vision Transformers po...
Tremendous amount of data are being generated and saved in many complex engineering and social systems every *** is significant and feasible to utilize the big data to make better decisions by machine learning techniq...
详细信息
Tremendous amount of data are being generated and saved in many complex engineering and social systems every *** is significant and feasible to utilize the big data to make better decisions by machine learning techniques. In this paper, we focus on batch reinforcement learning(RL) algorithms for discounted Markov decision processes(MDPs) with large discrete or continuous state spaces, aiming to learn the best possible policy given a fixed amount of training data. The batch RL algorithms with handcrafted feature representations work well for low-dimensional MDPs. However, for many real-world RL tasks which often involve high-dimensional state spaces, it is difficult and even infeasible to use feature engineering methods to design features for value function approximation. To cope with high-dimensional RL problems, the desire to obtain data-driven features has led to a lot of works in incorporating feature selection and feature learning into traditional batch RL algorithms. In this paper, we provide a comprehensive survey on automatic feature selection and unsupervised feature learning for high-dimensional batch RL. Moreover, we present recent theoretical developments on applying statistical learning to establish finite-sample error bounds for batch RL algorithms based on weighted Lpnorms. Finally, we derive some future directions in the research of RL algorithms, theories and applications.
This paper mainly focuses on designing a sliding mode boundary controller for a single flexible-link manipulator based on adaptive radial basis function (RBF) neural network. The flexible manipulator in this paper i...
详细信息
This paper mainly focuses on designing a sliding mode boundary controller for a single flexible-link manipulator based on adaptive radial basis function (RBF) neural network. The flexible manipulator in this paper is considered to be an Euler-Bernoulli beam. We first obtain a partial differential equation (PDE) model of single-link flexible manipulator by using Hamiltons approach. To improve the control robustness, the system uncertainties including modeling uncertainties and external disturbances are compensated by an adaptive neural approximator. Then, a sliding mode control method is designed to drive the joint to a desired position and rapidly suppress vibration on the beam. The stability of the closed-loop system is validated by using Lyapunov's method based on infinite dimensional model, avoiding problems such as control spillovers caused by traditional finite dimensional truncated models. This novel controller only requires measuring the boundary information, which facilitates implementation in engineering practice. Favorable performance of the closed-loop system is demonstrated by numerical simulations.
Background: This study aimed to investigate the feasibility and clinical benefits of indocyanine green (ICG) inhalation for detecting air leak sites during video-assisted thoracoscopic surgery (VATS). Methods: Between...
详细信息
Group behavior forecasting is an emergent re- search and application field in social computing. Most of the existing group behavior forecasting methods have heavily re- lied on structured data which is usually hard to...
详细信息
Group behavior forecasting is an emergent re- search and application field in social computing. Most of the existing group behavior forecasting methods have heavily re- lied on structured data which is usually hard to obtain. To ease the heavy reliance on structured data, in this paper, we pro- pose a computational approach based on the recognition of multiple plans/intentions underlying group behavior. We fur- ther conduct human experiment to empirically evaluate the effectiveness of our proposed approach.
DO we need a fundamental change in our professional culture and knowledge foundation for control and automation?If so,what are necessary and critical steps we must take to ensure such a change would take place effecti...
详细信息
DO we need a fundamental change in our professional culture and knowledge foundation for control and automation?If so,what are necessary and critical steps we must take to ensure such a change would take place effectively and efficiently,or more general,smoothly and sustainably?
Accurate detection of malignant transformation in oral potentially malignant disorders (OPMDs) is crucial for guiding effective treatment and improving patient management. This study evaluates the potential of MET- bi...
详细信息
Accurate detection of malignant transformation in oral potentially malignant disorders (OPMDs) is crucial for guiding effective treatment and improving patient management. This study evaluates the potential of MET- binding peptide-indocyanine green (cMBP-ICG), a mesenchymal-epithelial transition factor (MET)-targeted near-infrared fluorescence imaging (NIRFI) probe, for biopsy site selection in OPMDs. Preclinical results demonstrate the superior accuracy of NIRFI-assisted biopsy over conventional oral examination (COE)based biopsy in detecting high-grade dysplasia (HGD) or squamous cell carcinoma (SCC) and reducing missed detection rates. In a clinical trial with 50 patients, NIRFI-assisted biopsy achieves significantly higher diagnostic accuracy compared to COE-based biopsy (91% vs. 72%, p = 0.0005). These findings underscore the importance of NIRFI in enhancing diagnostic precision, supporting early detection and enabling timely and accurate treatment interventions for patients with OPMDs. The clinical trial is registered under the registration number ChiCTR2300074454.
View variation is a major challenge in face recognition. In this study, the authors propose a novel cross-view face recognition method by seeking potential intermediate domains between the source and target views to m...
详细信息
View variation is a major challenge in face recognition. In this study, the authors propose a novel cross-view face recognition method by seeking potential intermediate domains between the source and target views to model the connection of varying-views faces. Specifically, each intermediate domain is associated with a dictionary subspace. Learning proceeds in two phases. First, the authors discriminatively train a sub-dictionary for each subclass of data, which then compose a structured dictionary of powerful reconstructive and discriminative capability on the source data. Secondly, the authors gradually adapt the source domain dictionary to the target domain by incrementally reducing the reconstruction error on the target data, which forms a smooth transition path connecting the source and target domains. Instead of updating the structured dictionary integrally, the authors develop a refined sub-dictionary-based updating algorithm, which makes the intermediate dictionaries fit on the target data better and faster. Finally, the authors apply invariant sparse codes across the source, intermediate and target domains to render domain-shared representations, where the sample differences caused by view changes are reduced. Experiments on the CMU-PIE and Multi-PIE dataset demonstrate the effectiveness of the proposed method.
Recently, attributes have been introduced as a kind of high-level semantic information to help improve the classification accuracy. Multitask learning is an effective methodology to achieve this goal, which shares low...
详细信息
Recently, attributes have been introduced as a kind of high-level semantic information to help improve the classification accuracy. Multitask learning is an effective methodology to achieve this goal, which shares low-level features between attributes and actions. Yet such methods neglect the constraints that attributes impose on classes, which may fail to constrain the semantic relationship between the attributes and actions. In this paper, we explicitly consider such attribute-action relationship for human action recognition, and correspondingly, we modify the multitask learning model by adding attribute regularization. In this way, the learned model not only shares the low-level features, but also gets regularized according to the semantic constrains. In addition, since attribute and class label contain different amounts of semantic information, we separately treat attribute classifiers and action classifiers in the framework of multitask learning for further performance improvement. Our method is verified on three challenging datasets (KTH, UIUC, and Olympic Sports), and the experimental results demonstrate that our method achieves better results than that of previous methods on human action recognition.
暂无评论