A5003494611- Reinforcement Learning in Robotics
- Evolutionary Algorithms and Applications
- Robot Manipulation and Learning
- Machine Learning and Data Classification
- Telemedicine and Telehealth Implementation
- Robotic Mechanisms and Dynamics
- Connexins and lens biology
- Domain Adaptation and Few-Shot Learning
- Speech Recognition and Synthesis
- Analytical Chemistry and Chromatography
- Music and Audio Processing
- Metabolomics and Mass Spectrometry Studies
- BIM and Construction Integration
- Bee Products Chemical Analysis
- Mass Spectrometry Techniques and Applications
- Scheduling and Optimization Algorithms
- Teleoperation and Haptic Systems
- Hormonal and reproductive studies
- Speech and Audio Processing
University of Science and Technology of China
2019-2024
Fudan University
2021
Imaging of cholesterol and other metabolites simultaneously by ambient mass spectrometry will greatly benefit biological studies, however, it still remains challenging. Herein, adding acid into the desorption electrospray ionization (DESI) spray solvent, we achieved simultaneous imaging directly from mouse brain sections. The introduction increased signal intensity in tissues approximately 21-fold. Additionally, present strategy provided intensities for up to 62-fold, as well identification...
This study investigates the inhibitory effect of astaxanthin (AST) on testosterone-induced benign prostatic hyperplasia (BPH) in rats. Except for sham operation, BPH model rats were randomly assigned to five groups: control rats, AST-treated (20 mg/kg, 40 and 80 mg/kg), epristeride (EPR)-treated After treatment, as compared with prostate ventral weights decreased, while there was a marked decline mg/kg The same also observed index index. proliferation characteristics epithelia group...
Offline reinforcement learning (RL) has shown great potential in many robotic tasks, where doing trial-and-error with the environment is risky, costly, or time-consuming. However, it still hard to succeed long-horizon tasks especially when given suboptimal and multimodal offline datasets. Nevertheless, existing RL methods rarely consider structured information datasets, which are commonly found tasks. To address these challenges, we propose a novel approach that combines techniques of...
Offline reinforcement learning (RL) provides a promising approach to avoid costly online interaction with the real environment. However, performance of offline RL highly depends on quality datasets, which may cause extrapolation error in process. In many robotic applications, an inaccurate simulator is often available. data directly collected from cannot be used due well-known exploration-exploitation dilemma and dynamic gap between simulation To address these issues, we propose novel...
Dexterous grasping is a fundamental yet challenging skill in robotic manipulation, requiring precise interaction between hands and objects. In this paper, we present D(R,O) Grasp, novel framework that models the hand its pose object, enabling broad generalization across various robot object geometries. Our model takes hand's description point cloud as inputs efficiently predicts kinematically valid stable grasps, demonstrating strong adaptability to diverse embodiments Extensive experiments...
Dexterous manipulation is a critical area of robotics. In this field, teleoperation faces three key challenges: user-friendliness for novices, safety assurance, and transferability across different platforms. While collecting real robot dexterous data by to train robots has shown impressive results on diverse tasks, due the morphological differences between human hands, it not only hard new users understand action mapping but also raises potential concerns during operation. To address these...