Article
Environmental Sciences
Yuchen Guo, Lan Du, Guoxin Lyu
Summary: This study proposes a domain adaptive Faster R-CNN approach for SAR target detection with small training data size, transferring knowledge from labeled optical remote sensing images to SAR images to achieve higher detection accuracy.
Article
Neurosciences
Rakhi Agarwal, Asif Hussain, S. K. M. Varadhan, Domenico Campolo
Summary: Motor learning is an essential aspect of human behavior, and this study proposes a performance-based adaptive algorithm for task difficulty variation. The algorithm successfully adjusted the task difficulty based on user performance, leading to significant improvements in postural error after training.
FRONTIERS IN HUMAN NEUROSCIENCE
(2022)
Article
Chemistry, Multidisciplinary
Qianqian Yu, Keqi Fan, Yiyang Wang, Yuhui Zheng
Summary: With the rapid development of deep learning techniques, this study proposes a novel model, Faster MDNet, for object tracking. The model incorporates a channel attention module and domain adaptation components to improve tracking accuracy, while reducing model complexity and accelerating tracking speed with an adaptive spatial pyramid pooling layer.
APPLIED SCIENCES-BASEL
(2022)
Article
Computer Science, Artificial Intelligence
Jingjing Li, Mengmeng Jing, Hongzu Su, Ke Lu, Lei Zhu, Heng Tao Shen
Summary: This paper addresses the problem of accelerating machine learning in situations where there is a lack of training data and limited computing power. By proposing the Faster Domain Adaptation (FDA) protocol and two paradigms, the method achieves comparable or even better accuracy while using fewer computing resources.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2022)
Article
Computer Science, Artificial Intelligence
Alberto Torres-Barran, Carlos M. Alaiz, Jose R. Dorronsoro
Summary: The study proposes an improved version of the SMO algorithm based on Conjugate Descent procedure, which reduces the number of iterations while maintaining computational efficiency, especially with a positive definite kernel matrix leading to linear convergence rate. Experimental results demonstrate that this new approach is faster for many hyper-parameter configurations compared to traditional algorithms.
PATTERN RECOGNITION
(2021)
Article
Computer Science, Artificial Intelligence
Zhenwei He, Lei Zhang, Xinbo Gao, David Zhang
Summary: In this study, a Multi-Adversarial FasterRCNN (MAF) framework is proposed to address the cross-domain object detection task. The framework introduces Hierarchical Domain Feature Alignment (HDFA) and Aggregated Proposal Feature Alignment (APFA) modules to reduce domain disparities and improve detection performance. Furthermore, a Paradigm Teacher MAF (PT-MAF) framework is proposed with knowledge distillation and DualDiscriminator HDFA (D2-HDFA) modules to enhance domain adaptability and alignment.
INTERNATIONAL JOURNAL OF COMPUTER VISION
(2023)
Article
Computer Science, Artificial Intelligence
Lin Xiong, Mao Ye, Dan Zhang, Yan Gan, Yiguang Liu
Summary: In this paper, a source data-free domain adaptive object detection method is proposed. By utilizing unlabeled target domain data to optimize the source domain model, the method introduces a prototype-guided approach to extract semantic category information and iteratively updating global class prototypes to handle class and sample imbalances. Furthermore, a more accurate pseudo-label generation method is developed by combining semantic and image information.
PATTERN RECOGNITION
(2022)
Article
Neurosciences
Sarah A. Wilterson, Jordan A. Taylor
Summary: Learning in sensorimotor adaptation tasks has been traditionally seen as implicit learning, but recent studies have shown constraints in this process, questioning its effectiveness and application in motor learning paradigms. After five consecutive days of training, implicit learning plateaued without reaching full adaptation, particularly in mirror-reversal tasks.
Article
Neurosciences
Michele Deantoni, Thomas Villemonteix, Evelyne Balteau, Christina Schmidt, Philippe Peigneux
Summary: Functional magnetic resonance imaging (fMRI) studies have shown that offline consolidation of memory relies on continued neural activity during sleep, with sleep deprivation leading to less efficient memory consolidation. Resting state fMRI reveals that navigation-related activity persists during sleep and that sleep quality impacts the consolidation of memory traces and the brain's use of resources for linking spatial information.
Article
Computer Science, Artificial Intelligence
Hengtao Zhang, Zebin Yang, Agus Sudjianto, Aijun Zhang
Summary: This article proposes a novel sequential approach based on Stein's lemma for training additive index models (AIMs). The proposed SeqStein method successfully decouples the training of AIMs into two separable steps: Stein's estimation of the projection indices and nonparametric estimation of ridge functions using smoothing splines. Numerical experiments show that the SeqStein algorithm is not only more efficient for training AIMs, but also inclined to produce interpretable models with smooth ridge functions and sparse, nearly orthogonal projection indices.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Neurosciences
Evangelia-Regkina Symeonidou, Daniel P. Ferris
Summary: Intermittent visual occlusions can enhance balance training, leading to significant improvement in walking on a narrow balance beam.
FRONTIERS IN HUMAN NEUROSCIENCE
(2022)
Article
Neurosciences
Hee Yeon Im, Joshua J. Liddy, Joo-Hyun Song
Summary: This study investigates the impact of attentional context on relearning and recall in visuomotor adaptation. The findings suggest that the attentional context plays a crucial role in these processes, independent of explicit awareness of perturbations or changes in secondary task requirements.
JOURNAL OF NEUROPHYSIOLOGY
(2022)
Article
Neurosciences
Susen Werner, Koki Hasegawa, Kazuyuki Kanosue, Heiko K. Strueder, Tobias Goeb, Tobias Vogt
Summary: Recent research showed that there is no significant difference in the explicit intermanual transfer between judokas and running experts, but there is a difference in implicit intermanual transfer. The total intermanual transfer is mainly influenced by awareness of the perturbation, while implicit intermanual transfer is affected by factors such as judo training, total training volume, adaptation speed, and handedness scores.
EUROPEAN JOURNAL OF NEUROSCIENCE
(2021)
Article
Computer Science, Artificial Intelligence
Yong Zhang, Chaoxu Mu, Dongbin Zhao
Summary: This article proposes a data generation feedback relearning (DGFR) control algorithm, which improves the control performance by interacting with a delayed neural network-based data generation model. It avoids the security risk and poor real-time performance limitations of reinforcement learning algorithms.
IEEE TRANSACTIONS ON EMERGING TOPICS IN COMPUTATIONAL INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Yuhua Chen, Haoran Wang, Wen Li, Christos Sakaridis, Dengxin Dai, Luc Van Gool
Summary: This study introduces a model using adversarial training to learn domain classifiers, aiming to improve the robustness of object detection models across different domains. The model enhances performance by addressing domain shifts at two levels (image-level and instance-level).
INTERNATIONAL JOURNAL OF COMPUTER VISION
(2021)
Article
Behavioral Sciences
Laura A. Malone, Amy J. Bastian
NEUROBIOLOGY OF LEARNING AND MEMORY
(2016)
Article
Neurosciences
Rami J. Hamzey, Eileen M. Kirk, Erin V. L. Vasudevan
EXPERIMENTAL BRAIN RESEARCH
(2016)
Review
Neurosciences
E. P. Zehr, Trevor S. Barss, Katie Dragert, Alain Frigon, Erin V. Vasudevan, Carlos Haridas, Sandra Hundza, Chelsea Kaupp, Taryn Klarner, Marc Klimstra, Tomoyoshi Komiyama, Pamela M. Loadman, Rinaldo A. Mezzarane, Tsuyoshi Nakajima, Gregory E. P. Pearcey, Yao Sun
EXPERIMENTAL BRAIN RESEARCH
(2016)
Editorial Material
Pediatrics
Laura A. Malone, Anirudh Ramesh, Janet R. Serwint
PEDIATRICS IN REVIEW
(2017)
Article
Multidisciplinary Sciences
Erin V. L. Vasudevan, Rami J. Hamzey, Eileen M. Kirk
JOVE-JOURNAL OF VISUALIZED EXPERIMENTS
(2017)
Article
Multidisciplinary Sciences
Erin V. Vasudevan, Susan K. Patrick, Jaynie F. Yang
Review
Clinical Neurology
Laura A. Malone, Ryan J. Felling
PEDIATRIC NEUROLOGY
(2020)
Review
Clinical Neurology
Laura A. Malone, Lisa R. Sun
CURRENT TREATMENT OPTIONS IN NEUROLOGY
(2019)
Article
Neurosciences
Rania Almajid, Carole Tucker, Emily Keshner, Erin Vasudevan, William Geoffrey Wright
Summary: This study investigated the impact of wearing HMD on kinematic measures in younger and older adults, finding that wearing HMD decreased turning cadence and trunk velocities, and increased the time to complete tasks, especially in older adults. This suggests that the effects of HMD on physical performance should be considered in VR rehabilitation applications.
Article
Chemistry, Analytical
Siyao Hu, Krista Fjeld, Erin V. Vasudevan, Katherine J. Kuchenbecker
Summary: The new device for gait rehabilitation, the Gait Propulsion Trainer (GPT), applies periodic stance-phase resistance to improve walking patterns. Testing with healthy individuals and stroke survivors showed different responses, with promising short-term results for hemiparetic gait. Further studies will explore long-term effects.
Article
Neurosciences
Dulce M. Mariscal, Erin V. L. Vasudevan, Laura A. Malone, Gelsy Torres-Oviedo, Amy J. Bastian
Summary: Humans have the ability to adapt and correct movement errors based on changes in the body and environment. This study suggests that the ability to adapt locomotor patterns in different environments decreases with age, indicating that age and experience play crucial roles in regulating the specificity of motor learning.
Article
Rehabilitation
Laura A. Malone, Amanda Morrow, Yuxi Chen, Donna Curtis, Sarah D. de Ferranti, Monika Desai, Talya K. Fleming, Therese M. Giglia, Trevor A. Hall, Ellen Henning, Sneha Jadhav, Alicia M. Johnston, Dona Rani C. Kathirithamby, Christina Kokorelis, Catherine Lachenauer, Lilun Li, Henry C. Lin, Tran Locke, Carol MacArthur, Michelle Mann, Sharon A. McGrath-Morrow, Rowena Ng, Laurie Ohlms, Sarah Risen, S. Christy Sadreameli, Sarah Sampsel, S. Kristen Sexson Tejtel, Julie K. Silver, Tregony Simoneau, Rasha Srouji, Sanjeev Swami, Souraya Torbey, Monica Verduzco Gutierrez, Cydni Nicole Williams, Lori Allison Zimmerman, Louise Elaine Vaz
Review
Clinical Neurology
Nayo M. Hill, Laura A. Malone, Lisa R. Sun
Summary: Pediatric stroke, which causes static damage to the developing brain, has unique implications on motor, sensory, cognitive, speech, and behavioral functions. Timing and location of stroke, as well as comorbid conditions, play a significant role in outcomes.
PEDIATRIC NEUROLOGY
(2023)
Article
Physiology
Laura A. A. Malone, Nayo M. M. Hill, Haley Tripp, Daniel M. M. Wolpert, Amy J. J. Bastian
Summary: We designed a remote game that could be played at home through a web browser. The game required children to adapt to a visuomotor rotation between their hand movement and a ball displayed in the game. The task had several novel features that allowed the study of adaptation across different ages. We tested the concurrent validity of the remote task by comparing it to the same task performed in a laboratory.
PHYSIOLOGICAL REPORTS
(2023)
Article
Neurosciences
Matthew A. Statton, Alejandro Vazquez, Susanne M. Morton, Erin V. L. Vasudevan, Amy J. Bastian