A5101741369- Reinforcement Learning in Robotics
- Neural Networks and Applications
- AI-based Problem Solving and Planning
- Robot Manipulation and Learning
- Photonic and Optical Devices
- Software System Performance and Reliability
- Hand Gesture Recognition Systems
- Analytical Chemistry and Sensors
- Computational Physics and Python Applications
- 3D Shape Modeling and Analysis
- Neural Networks and Reservoir Computing
- Online Learning and Analytics
- Mechanical and Optical Resonators
- Advanced Vision and Imaging
- Random lasers and scattering media
- Geological Modeling and Analysis
- Drilling and Well Engineering
- Distributed and Parallel Computing Systems
- Software Testing and Debugging Techniques
- Explainable Artificial Intelligence (XAI)
- Data Mining Algorithms and Applications
- Seismic Imaging and Inversion Techniques
- Silicon Nanostructures and Photoluminescence
- Software Reliability and Analysis Research
- Computer Graphics and Visualization Techniques
Linköping University
2024
Amazon (United States)
2024
University of California, Los Angeles
2010-2019
Shanghai Normal University
2018
Lawrence Berkeley National Laboratory
2013
Forms of explanation that are best suited to foster trust do not necessarily correspond those components contributing the task performance.
As 3D content becomes increasingly prevalent, there's a growing focus on the development of engagements with virtual content. Unfortunately, traditional techniques for creating, editing, and interacting this are fraught difficulties. They tend to be not only engineering-intensive but also require extensive expertise, which adds frustration inefficiency in object manipulation. Our proposed VR-GS system represents leap forward human-centered interaction, offering seamless intuitive user...
We present a design of an easy-to-replicate glove-based system that can reliably perform simultaneous hand pose and force sensing in real time, for the purpose collecting human data during fine manipulative actions. The consists sensory glove is capable jointly finger poses, as well forces on palm each phalanx. Specifically, employs network 15 IMUs to measure rotations between individual phalanxes. Hand then reconstructed using forward kinematics. Contact phalanx are measured by 6 customized...
This paper presents a microscale resonant sensor that has been fabricated with nanoscale pores for enhanced sensitivity to chemical vapors. By building resonators are made of porous silicon, we take advantage the increased area molecular binding and improve vapor concentration interest. We present results whose surfaces entirely silicon. also examine use targeted regions porosity keep critical parts beam nonporous mechanically stable while still maximizing surface area. Surface...
In the field of cryptography, it is crucial that random numbers used in key generation are not only genuinely but also private, meaning no other party than legitimate user must have information about generated. Quantum number generators can offer both properties - fundamentally output, as well ability to implement certify amount private randomness generated, order remove some side-channel attacks. this study we introduce perovskite technology a resilient platform for photonics, where...
Multi-task learning (MTL) is a paradigm that simultaneously learns multiple tasks by sharing information at different levels, enhancing the performance of each individual task. While previous research has primarily focused on feature-level or parameter-level task relatedness, and proposed various model architectures algorithms to improve performance, we aim explore output-level relatedness. This approach introduces posteriori into model, considering may produce correlated outputs with mutual...
Aimed at the GUI (graphical user interface) software testing existence of complexity and inefficiency issues, propose an Event Interactive Graph tree model. The structural events system interactive separate to establish component graph event model, gives a detailed modeling methods, according model designed test coverage criteria cases generation method. Finally applied achieve good results in actual has greatly enhanced quality efficiency testing.
Miniaturization of mass-based resonant sensors is a promising strategy for chemical detection due to the increased sensitivity afforded by decreased length scales. However, while surface area-to-volume ratios microand nano-scale devices provide benefits, smaller scales pose additional challenges in fabrication. We introduce use porous silicon resonators vapor with minimal fabrication challenge. Standard top-down processes were used fabricate micrometer-scale nanometerscale pores. The area...
In searching for a generalizable representation of temporally extended tasks, we spot two necessary constituents: the utility needs to be non-Markovian transfer temporal relations invariant probability shift, also lifted abstract out specific grounding objects. this work, study learning such from human demonstrations. While inverse reinforcement (IRL) has been accepted as general framework learning, its fundamental formulation is one concrete Markov Decision Process. Thus learned reward...