There is clear evidence indicating that the standardized Sensing-Based Semi-Persistent Scheduling scheme in Out-of-coverage possesses specific vulnerabilities. This unsatisfactory performance lies in the absence of co...
详细信息
ISBN:
(数字)9798350370539
ISBN:
(纸本)9798350370546
There is clear evidence indicating that the standardized Sensing-Based Semi-Persistent Scheduling scheme in Out-of-coverage possesses specific vulnerabilities. This unsatisfactory performance lies in the absence of coordination in channel scheduling and transmission time. Specifically, the determination to re-assign a resource after the reservation period expires is currently based on a random selection process, without consideration of the actual interference encountered by the utilized resources. In this study, a novel scheme is proposed to markedly improve the random selection method for vehicle positioning.
Unilateral muscle weakness and hand paralysis is the most common outcome after a stroke. Functional electrical stimulation (FES) is effective in assisting hand prehension, but conventional stimulators require users to...
详细信息
Unilateral muscle weakness and hand paralysis is the most common outcome after a stroke. Functional electrical stimulation (FES) is effective in assisting hand prehension, but conventional stimulators require users to manually select the grasp type (e.g. by pressing a button), which is challenging for patients with severe paralysis. Also, patients need to frequently divert attention from their task to operate the stimulator, which is cumbersome and reduces their engagement in the therapy. In this study, we develop a novel deep learning-based object detection approach to select multiple grasp types and control an electrical stimulator, in order to assist grasping. Object detection was performed using a state-of-the-art YOLOv5 algorithm, which achieved above 93% mean average precision. The algorithm tracked the positions of the hand and objects and selected a grasp type based on the object nearest to the hand. Once the grasp type was selected, a custom-built FES stimulator was activated to execute pre-defined stimulation sequences and allow a person to grasp the nearest object. This contactless, vision-based solution is beneficial for patients opting for homebased rehabilitation since it doesn’t require additional setup time or help from caregivers. The future scope of this work includes testing the object detection-based FES on stroke patients and determining its efficacy in restoring hand movements.
This study examines to what extent the testing of traditional software components and machine learning (ML) models fundamentally differs or not. While some researchers argue that ML software requires new concepts and ...
详细信息
This study examines to what extent the testing of traditional software components and machine learning (ML) models fundamentally differs or not. While some researchers argue that ML software requires new concepts and perspectives for testing, our analysis highlights that, at a fundamental level, the specification and testing of a software component are not dependent on the development process used or on implementation details. Although the software engineering/computerscience (SE/CS) and Data science/ML (DS/ML) communities have developed different expectations, unique perspectives, and varying testing methods, they share clear commonalities that can be leveraged. We argue that both areas can learn from each other, and a non-dual perspective could provide novel insights not only for testing ML but also for testing traditional software. Therefore, we call upon researchers from both communities to collaborate more closely and develop testing methods and tools that can address both traditional and ML software components. While acknowledging their differences has merits, we believe there is great potential in working on unified methods and tools that can address both types of software.
Online knowledge distillation (KD) has drawn increasing attention in recent years. However, little attention has been paid to the capacity gap that exists between the models in the online KD paradigm. In this work, we...
Online knowledge distillation (KD) has drawn increasing attention in recent years. However, little attention has been paid to the capacity gap that exists between the models in the online KD paradigm. In this work, we investigate the impact of the capacity gap and experimentally verified that a large capacity gap can have a detrimental effect on performance. Moreover, to address this issue, we propose Auxiliary Branch assisted Mutual Learning (ABML), which leverages auxiliary branches to mitigate performance deterioration caused by the capacity gap. Experimental results show that our ABML can significantly improve the performance of online KD even when a capacity gap is present. Our code is available at https://***/maorongwang/ABML.
Active learning is a powerful technique that accelerates model learning by iteratively expanding training data based on the model's feedback. This approach has proven particularly relevant in natural language proc...
Active learning is a powerful technique that accelerates model learning by iteratively expanding training data based on the model's feedback. This approach has proven particularly relevant in natural language processing and other machine learning domains. While active learning has been extensively studied for conventional classification tasks, its application to more specialized tasks like neural coreference resolution has the potential for improvement. In our research, we present a significant advancement by applying active learning to the neural coreference problem, and setting a benchmark of 39 % reduction in required annotations for training data. Simultaneously, it preserves performance compared to the original model trained on the full data. We compare various uncertainty sampling techniques along with Bayesian modifications of coreference resolution models, conducting a comprehensive analysis of annotation efforts. The results demonstrate that the best-performing techniques seek to maximize label annotation in previously chosen documents, showcasing their effectiveness and preserving performance.
The risen complexity of automotive systems requires new development strategies and methods to master the upcoming challenges. Traditional methods need thus to be changed by an increased level of automation, and a fast...
详细信息
Embedded index coding, known as device-to-device index coding, is a promising transmission scheme that effectively reduces the number of transmissions required in device-to-device communications. Leveraging spatial mu...
Embedded index coding, known as device-to-device index coding, is a promising transmission scheme that effectively reduces the number of transmissions required in device-to-device communications. Leveraging spatial multiplexing gain attained with multiple antennas in wireless environment, this paper explores the benefits of spatially multiplexed embedded index codes through the joint design of embedded index coding and multicast beamformers. Through our simulations, we are able to demonstrate significant improvements obtained by the proposed scheme in terms of spatially multiplexed embedded index coding.
The paper discusses low-cost design closure of antenna structures. Our approach exploits trust-region (TR) gradient search with antenna response gradients updated by a sparse application of finite differentiation, lim...
详细信息
This paper presents the research activities in the context of the SPADES project for scalable indexing and processing of big spatial and spatio-textual data. Management of spatio-textual data raises challenges due to ...
详细信息
The use of asserts in code has received increasing attention in the software engineering community in the past few years, even though it has been a recognized programming construct for many decades. A previous empiric...
详细信息
暂无评论