A5101464065- Quantum Information and Cryptography
- Quantum Computing Algorithms and Architecture
- Neural Networks and Reservoir Computing
- Topic Modeling
- Advanced Memory and Neural Computing
- Neural Networks and Applications
- Natural Language Processing Techniques
- Quantum and electron transport phenomena
- Multimodal Machine Learning Applications
- Text and Document Classification Technologies
- Domain Adaptation and Few-Shot Learning
- Advanced Text Analysis Techniques
- Advanced Graph Neural Networks
- Statistical Mechanics and Entropy
- Data Quality and Management
- Optical Network Technologies
- Philosophy and History of Science
- Gaussian Processes and Bayesian Inference
- Lattice Boltzmann Simulation Studies
- Face recognition and analysis
- Biometric Identification and Security
- Fluid Dynamics and Turbulent Flows
- Advanced Optimization Algorithms Research
- Metaheuristic Optimization Algorithms Research
- Educational Technology and Pedagogy
Xinjiang University
2022-2025
Queensland Rail
2024
Peking University
2020-2023
Zhejiang Sci-Tech University
2023
Fudan University
2023
Ningbo University
2022
UNSW Sydney
2019-2021
The University of Sydney
2021
Q-Flex (United States)
2021
ARC Centre of Excellence for Engineered Quantum Systems
2021
As interest in quantum computing grows, there is a pressing need for standardized API's so that algorithm designers, circuit and physicists can be provided common reference frame designing, executing, optimizing experiments. There also language specification goes beyond gates allows users to specify the time dynamics of experiment recover output. In this document we provide interface backends (simulators experiments) standarized data structure (Qobj --- object) sending experiments those via...
Effectively manipulating quantum computing hardware in the presence of imperfect devices and control systems is a central challenge realizing useful computers.Susceptibility to noise critically limits performance capabilities today's so-called noisy intermediate-scale (NISQ) devices, as well any future technologies.Fortunately, enables efficient execution logic operations algorithms with builtin robustness errors, without need for complex logical encoding.In this manuscript we introduce...
The combination of machine learning and quantum computing has emerged as a promising approach for addressing previously untenable problems. Reservoir is an efficient paradigm that utilizes nonlinear dynamical systems temporal information processing, i.e., processing input sequences to produce output sequences. Here we propose reservoir harnesses complex dissipative dynamics. Our class reservoirs universal, in any fading memory map can be approximated arbitrarily closely uniformly over all...
The dynamic expansion architecture is becoming popular in class incremental learning, mainly due to its advantages alleviating catastrophic forgetting. However, task confu- sion not well assessed within this framework, e.g., the discrepancy between classes of different tasks learned (i.e., inter-task confusion, ITC), and certain prior- ity still given latest batch old-new con- fusion, ONC). We empirically validate side effects two types confusion. Meanwhile, a novel solution called Task...
Identifying and calibrating quantitative dynamical models for physical quantum systems is important a variety of applications. Here we present closed-loop Bayesian learning algorithm estimating multiple unknown parameters in model, using optimized experimental ``probe'' controls measurement. The estimation based on particle filter, designed to autonomously choose informationally probe experiments with which compare model predictions. We demonstrate the performance both simulated calibration...
Slender-body aircraft operating at high angles of attack often experience nonlinear, asymmetric multi-vortex flow structures that generate random, unsteady lateral forces, undermining stability, and maneuverability. Dielectric barrier discharge plasma actuators can eliminate these forces. However, conventional open-loop control method cannot adapt to dynamic fields in real time, limiting the overall effectiveness active control. This study introduces a framework grounded physical principles...
Named Entity Recognition (NER) aims to identify entities with specific meanings and their boundaries in natural language texts. Due the differences between Chinese English families, NER faces challenges such as ambiguous word boundary delineation semantic diversity. Previous studies on have focused character lexical information, neglecting unique feature of Chinese—pinyin information. In this paper, we propose CPL-NER, which combines multiple information characters embedding enhance...
Contrastive learning (CL) has been successfully applied in Natural Language Processing (NLP) as a powerful representation method and shown promising results various downstream tasks. Recent research highlighted the importance of constructing effective contrastive samples through data augmentation. However, current augmentation methods primarily rely on random word deletion, substitution, cropping, which may introduce noisy hinder learning. In this article, we propose novel approach to...
Based on a recently developed theory, we propose realization of single-input single-output (SISO) quantum reservoir computer near-term for approximating SISO fading memory maps. Such can be interest applications such as nonlinear system identification and signal processing (e.g., speech processing), opening an avenue early control-oriented computers. We detail implementation the cloud-based 20 qubit IBM report simulation results proposed scheme Qiskit simulator this machine.
Nonlinear stochastic modeling is useful for describing complex engineering systems. Meanwhile, neuromorphic (brain-inspired) computing paradigms are being developed to tackle tasks that challenging and resource-intensive on digital computers. An emerging scheme reservoir computing, which exploits nonlinear dynamical systems temporal information processing. This brief introduces computers with output feedback as stationary ergodic infinite-order autoregressive models. We highlight the...
Manipulating quantum computing hardware in the presence of imperfect devices and control systems is a central challenge realizing useful computers. Susceptibility to noise limits performance capabilities noisy intermediate-scale (NISQ) devices, as well any future technologies. Fortunately enables efficient execution logic operations algorithms with built-in robustness errors, without need for complex logical encoding. In this manuscript we introduce software tools application integration...
Face features, as the most widely adopted and essential biometric characteristic in identity verification recognition, play a crucial role ensuring security. However, significance is also accompanied by various face attacks, posing great threat to security of facial recognition systems. Therefore, exploring anti-spoofing detection holds substantial practical significance. There existing research on Anti-Spoofing (FAS) detection, but there still potential for enhancing performance. In this...
Identifying and calibrating quantitative dynamical models for physical quantum systems is important a variety of applications. Here we present closed-loop Bayesian learning algorithm estimating multiple unknown parameters in model, using optimised experimental "probe" controls measurement. The estimation based on particle filter, designed to autonomously choose informationally-optimised probe experiments with which compare model predictions. We demonstrate the performance both simulated...
Recent efforts to develop hybrid quantum-classical algorithms for solving combinatorial problems have rekindled interest in revisiting heuristic classical optimization and exploring possibilities improving them. A popular approach finding good solutions is local search. In spite of its efficiency, if the search space rugged, often gets trapped unsatisfactory optima. On other hand, global meta-heuristic algorithms, such as simulated annealing, guarantee asymptotic convergence probability...
In this paper, we propose a unified framework for an abstractive summarization method which uses the prompt language model and pointer mechanism. The problem usually includes text encoder decoder. Current methods employ encoder-decoder architecture to condense paraphrase document. To better document, that only topic-sensitive Our has input module, decoder We apply our Xsum, Gigaword, CNN/DailyMail datasets, experimental results demonstrate achieved state-of-the-art on Xsum dataset comparable...
Named Entity Recognition (NER) poses challenges for both flat and nested tasks, which require different paradigms. To overcome this issue, we present GFNER, a unified global feature-aware framework based on table filling, that can handle types of tasks with low computational cost. While pretrained models have shown great promise in NER, they typically focus local contextual information, disregarding relationships are crucial accurate entity boundary extraction. address limitation, introduce...