A5100359839- Face and Expression Recognition
- Topic Modeling
- Domain Adaptation and Few-Shot Learning
- Advanced Image and Video Retrieval Techniques
- Sparse and Compressive Sensing Techniques
- Natural Language Processing Techniques
- Anomaly Detection Techniques and Applications
- Advanced Causal Inference Techniques
- Advanced Graph Neural Networks
- Genomics and Phylogenetic Studies
- Web Data Mining and Analysis
- Recommender Systems and Techniques
- Remote-Sensing Image Classification
- RNA and protein synthesis mechanisms
- Machine Learning in Bioinformatics
- Advanced Text Analysis Techniques
- Image Retrieval and Classification Techniques
- Face recognition and analysis
- Advanced Computational Techniques and Applications
- Complex Network Analysis Techniques
- Bayesian Modeling and Causal Inference
- Statistical Methods and Inference
- Adversarial Robustness in Machine Learning
- Advanced Neural Network Applications
- Text and Document Classification Technologies
Dalian University of Technology
2013-2025
Wuhan Technical College of Communications
2023-2025
Nanjing University of Chinese Medicine
2023-2025
University of Virginia
2022-2025
Southwest Minzu University
2023-2025
Traditional Chinese Medicine Hospital of Kunshan
2025
Shanghai University of Traditional Chinese Medicine
2025
Huazhong Agricultural University
2025
Ministry of Agriculture and Rural Affairs
2025
PLA Army Engineering University
2025
This paper introduces McPAT, an integrated power, area, and timing modeling framework that supports comprehensive design space exploration for multicore manycore processor configurations ranging from 90nm to 22nm beyond. At the microarchitectural level, McPAT includes models fundamental components of a chip multiprocessor, including in-order out-of-order cores, networks-on-chip, shared caches, memory controllers, multiple-domain clocking. circuit technology levels, critical-path modeling,...
Collaborative filtering (CF) has been widely employed within recommender systems to solve many real-world problems. Learning effective latent factors plays the most important role in collaborative filtering. Traditional CF methods based upon matrix factorization techniques learn from user-item ratings and suffer cold start problem as well sparsity problem. Some improved enrich priors on by incorporating side information regularization. However, learned may not be very due sparse nature of...
Causal inference is a critical research topic across many domains, such as statistics, computer science, education, public policy, and economics, for decades. Nowadays, estimating causal effect from observational data has become an appealing direction owing to the large amount of available low budget requirement, compared with randomized controlled trials. Embraced rapidly developed machine learning area, various estimation methods have sprung up. In this survey, we provide comprehensive...
Scene graph generation has received growing attention with the advancements in image understanding tasks such as object detection, attributes and relationship prediction, etc. However, existing datasets are biased terms of labels, or often come noisy missing annotations, which makes development a reliable scene prediction model very challenging. In this paper, we propose novel algorithm external knowledge reconstruction loss to overcome these dataset issues. particular, extract commonsense...
Obtaining RNA-seq measurements involves a complex data analytical process with large number of competing algorithms as options. There is much debate about which these methods provides the best approach. Unfortunately, it currently difficult to evaluate their performance due in part lack sensitive assessment metrics. We present series statistical summaries and plots terms specificity sensitivity, available R/Bioconductor package ( http://bioconductor.org/packages/rnaseqcomp ). Using two...
<p>The sustainability of life on Earth is under increasing threat due to human-induced climate change. This perilous change in the Earth's caused by increases carbon dioxide and other greenhouse gases atmosphere, primarily emissions associated with burning fossil fuels. Over next two three decades, effects change, such as heatwaves, wildfires, droughts, storms, floods, are expected worsen, posing greater risks human health global stability. These trends call for implementation...
Text data augmentation is an effective strategy for overcoming the challenge of limited sample sizes in many natural language processing (NLP) tasks. This especially prominent few-shot learning scenario, where target domain generally much scarcer and lowered quality. A widely-used to mitigate such challenges perform better capture invariance increase size. However, current text methods either can't ensure correct labeling generated (lacking faithfulness) or sufficient diversity compactness),...
Multi-View Clustering (MVC) aims to find the cluster structure shared by multiple views of a particular dataset. Existing MVC methods mainly integrate raw data from different views, while ignoring high-level information. Thus, their performance may degrade due conflict between heterogeneous features and noises existing in each individual view. To overcome this problem, we propose novel Ensemble (MVEC) framework solve an (EC) way, which generates Basic Partitions (BPs) for view individually...
In this paper, we aim at learning robust and discriminative subspaces from noisy data. Subspace is widely used in extracting features for classification. However, when data are contaminated with severe noise, the performance of most existing subspace methods would be limited. Recent advances low-rank modeling provide effective solutions removing noise or outliers contained sample sets, which motivates us to take advantage constraints order exploit particular, present a method called...
The environment provides points for control of pathogens spread by food, water, hands, air, or fomites. These environmental transmission pathways require contact network formulations more realistically detailed than those based on social encounters physical proximity. As a step toward improved assessment interventions, description networks, and better use specimens to analyze transmission, an infection system model that describes the dynamics human interaction with in is presented. Its...
Phenomenally successful in practical inference problems, convolutional neural networks (CNN) are widely deployed mobile devices, data centers, and even supercomputers. The number of parameters needed CNNs, however, often large undesirable. Consequently, various methods have been developed to prune a CNN once it is trained. Nevertheless, the resulting CNNs offer limited benefits. While pruning fully connected layers reduces CNN's size considerably, does not improve speed noticeably as compute...
Image clustering has been a critical preprocessing step for vision tasks, e.g., visual concept discovery, content-based image retrieval. Conventional methods use handcraft descriptors as basic features via K-means, or build the graph within spectral clustering. Recently, representation learning with deep structure shows appealing performance in unsupervised feature pre-treatment. However, few studies have discussed how to deploy problems, especially unified framework which integrates both...
Multiview clustering (MVC), which aims to explore the underlying cluster structure shared by multiview data, has drawn more research efforts in recent years. To exploit complementary information among multiple views, existing methods mainly learn a common latent subspace or develop certain loss across different while ignoring higher level such as basic partitions (BPs) generated single-view algorithm. In light of this, we propose novel marginalized ensemble (M <sup...
Person re-identification plays an important role in many safety-critical applications. Existing works mainly focus on extracting patch-level features or learning distance metrics. However, the representation power of extracted might be limited, due to various viewing conditions pedestrian images complex real-world scenarios. To improve features, we learn discriminative and robust representations via dictionary this paper. First, propose a Cross-view Dictionary Learning (CDL) model, which is...
Visual data such as images and videos are easily accessible nowadays, they play critical roles in many real-world applications like surveillance. This raises a series of technological demands for automatic visual understanding content summarization, which has guided the research community to move towards better achievement capabilities. Meanwhile, it presents big challenge semantic video automatically translating them into human language. When developing translation systems, one issue is how...
Glycosyltransferases (GTs) are prevalent across the tree of life and regulate nearly all aspects cellular functions. The evolutionary basis for their complex diverse modes catalytic functions remain enigmatic. Here, based on deep mining over half million GT-A fold sequences, we define a minimal core component shared among functionally enzymes. We find that variations in common emergence hypervariable loops extending from contributed to diversity. provide phylogenetic framework relating...
As a method of function approximation, polynomial fitting has always been the main research hotspot in mathematical modeling. In many disciplines such as computer, physics, biology, neural networks have widely used, and most applications transformed into problems using networks. One reasons that can be used is it certain sense universal approximation. order to fit polynomial, this paper constructs three-layer feedforward network, uses Taylor series activation function, determines number...
Artificial general intelligence (AGI) has gained global recognition as a future technology due to the emergence of breakthrough large language models and chatbots such GPT-4 ChatGPT, respectively. Compared conventional AI models, typically designed for limited range tasks, demand significant amounts domain-specific data training may not always consider intricate interpersonal dynamics in education. AGI, driven by recent pre-trained represents leap capability machines perform tasks that...