A5102989092- Parallel Computing and Optimization Techniques
- Distributed systems and fault tolerance
- Advanced Manufacturing and Logistics Optimization
- Manufacturing Process and Optimization
- Target Tracking and Data Fusion in Sensor Networks
- Image and Object Detection Techniques
- Advanced Neural Network Applications
- Image Processing and 3D Reconstruction
- Astrophysics and Star Formation Studies
- Spectroscopy and Laser Applications
- Real-Time Systems Scheduling
- Domain Adaptation and Few-Shot Learning
- Topic Modeling
- Stellar, planetary, and galactic studies
- Software Engineering Research
- Neural Networks and Applications
- Time Series Analysis and Forecasting
- Scheduling and Optimization Algorithms
- Stochastic Gradient Optimization Techniques
- Generative Adversarial Networks and Image Synthesis
- Machine Learning and ELM
Tongji University
2023
University of Kansas
2023
Denso (United States)
2022
Convergence
2021
University of Chinese Academy of Sciences
2020
PLA Army Engineering University
2019
Over the past decade, image classification, which can provide assistance to address complex tasks such as planetary exploration and unmanned driving, has become a hot topic. As subproblem of scene classification received increasing attention. Based on previous studies, Xception model achieved superior performance in comparison with original Inception model. The is advantageous at processing yet it not been used for classification. To tackle this issue, paper proposed an based transfer...
Natural and formal languages provide an effective mechanism for humans to specify instructions reward functions. We investigate how generate policies via RL when functions are specified in a symbolic language captured by Reward Machines, increasingly popular automaton-inspired structure. interested the case where mapping of environment state (here, Machine) vocabulary -- commonly known as labelling function is uncertain from perspective agent. formulate problem policy learning Machines with...
Even though earliest-deadline-first (EDF) is optimal in terms of uniprocessor schedulability, it co-NP-hard to precisely verify schedulability for constrained-deadline task sets. The most efficient way solve this problem polynomial time via a partially linear approximation the demand bound function. Such leads simple testing with speedup factor ρ. result further Deadline-Monotonic Partitioned-EDF on multi-processors 1 + ρ − 1/m (where m number processors). current state art results indicate...
Soft errors occur frequently on large computing platforms due to the increasing scale and complexity of HPC systems. Various resilience techniques (e.g., checkpointing, ABFT, replication) have been proposed protect scientific applications from soft at different levels. Among them, system-level replication often involves duplicating or even triplicating entire computation, thus resulting in high overhead. This paper proposes dynamic selective protection for sparse iterative solvers,...
Aerospace components (ACs) are an important part of the aerospace equipment. Because process specialty, AC production scheduling has always been seen a challenging work. In this paper, improved dual-resource-constrained flexible job shop model is constructed to formulate problem. The characteristics manufacturing fully considered by analyzing relationship among processes, machines, and workers. Meanwhile, effect machine type, worker skill level labor efficiency also incorporated into model....
Zeeman effect characterizes the phenomenon that a spectral line is split into several components in presence of static magnetic field. It very important applications, such as resonance imaging, nuclear spectroscopy and atomic absorption spectroscopy. measurement was used Di Li's recent Nature paper to reveal coherent field from cold neutral medium molecular envelope. This proposes novel data observation based on neural network for detection, which provides different way learn feature...