Xin Zhang

Taming Convergence for Asynchronous Stochastic Gradient Descent with Unbounded Delay in Non-Convex Learning

2020

Understanding the convergence performance of asynchronous stochastic gradient descent method (Async-SGD) has received increasing attention in recent years due to their foundational role in machine learning. To date, however, most of the existing works are …

Influence of vapor-phase fluids on the geochemical characterization of hydrothermal sulfides in the shimmering waters of the southern Okinawa Trough

2021 · Ore Geology Reviews

Inverted lakes filled with shimmering water are unique environments within hydrothermal systems. Recently, vapor-phase fluids have been detected on the top of an inverted lake in a mushroom cap-shaped (MC) sulfide structure based on in …

FOA-based QR-factorized CKF Algorithm for AUV Navigation

2022 · OCEANS 2022, Hampton Roads

Navigation is one of the key technologies for autonomous underwater vehicles (AUVs). To improve the accuracy of the navigation system, it is necessary to select an appropriate algorithm to fuse the navigation parameters. This paper …

Intelligent recognition system of converter station operation text based on computer image acquisition

2023

In order to improve the efficiency of text recognition of converter station operation, this paper studies the intelligent text recognition system combined with computer image acquisition technology. Through the visual recognition algorithm developed by VS2010 …

REAF: Remembering Enhancement and Entropy-Based Asymptotic Forgetting for Filter Pruning

2023 · IEEE Transactions on Image Processing

Neurologically, filter pruning is a procedure of forgetting and remembering recovering. Prevailing methods directly forget less important information from an unrobust baseline at first and expect to minimize the performance sacrifice. However, unsaturated base remembering …

Towards Imbalanced Large Scale Multi-label Classification with Partially Annotated Labels

2023

Multi-label classification is a widely encountered problem in daily life, where an instance can be associated with multiple classes. In theory, this is a supervised learning method that requires a large amount of labeling. However, …

Weakly Supervised Video Anomaly Detection Based on Cross-Batch Clustering Guidance

2023

Weakly supervised video anomaly detection (WSVAD) is a challenging task since only video-level labels are available for training. In previous studies, the discriminative power of the learned features is not strong enough, and the data …

Efficient Sparse Least Absolute Deviation Regression With Differential Privacy

2024 · IEEE Transactions on Information Forensics and Security

In recent years, privacy-preserving machine learning algorithms have attracted increasing attention because of their important applications in many scientific fields. However, in the literature, most privacy-preserving algorithms demand learning objectives to be strongly convex and …

SpeechGPT-Gen: Scaling Chain-of-Information Speech Generation

2024 · arXiv (Cornell University)

Benefiting from effective speech modeling, current Speech Large Language Models (SLLMs) have demonstrated exceptional capabilities in in-context speech generation and efficient generalization to unseen speakers. However, the prevailing information modeling process is encumbered by certain …

IPOD2: an irrecoverable and verifiable deletion scheme for outsourced data

2024 · The Computer Journal

Abstract To alleviate the burden of data storage and management, there is a growing trend of outsourcing data to the cloud that enables users to remotely manage their data flexibly. However, this shift also raises …

CodeSwap: Symmetrically Face Swapping Based on Prior Codebook

2024

Face swapping, the technique of transferring the identity from one face to another, merges as a field with significant practical applications. However, previous swapping methods often result in visible artifacts. To address this issue, in …

A scientific-article key-insight extraction system based on multi-actor of fine-tuned open-source large language models

2025 · Scientific Reports

The exponential growth of scientific articles has presented challenges in information organization and extraction. Automation is urgently needed to streamline literature reviews and enhance insight extraction. We explore the potential of Large Language Models (LLMs) …

A risk assessment model for the entire rice processing chain based on Kmeans++ and extreme learning machine

2025 · LWT

The rice processing chain plays a crucial role in the rice supply chain, making the risk assessment of its main pollutants essential. Existing methods often rely on subjective weight determination and fail to consider the …

Deep learning-based data centre energy efficiency analysis and optimisation research

2025

With the rapid development of the digital economy, data centres, as the core carrier of the arithmetic infrastructure, are facing increasingly severe energy consumption problems. The traditional data centre operation is based on resource redundancy …

Deep Learning Modeling of Consumer Payment Behavior and Security Perception

2025 · Journal of Organizational and End User Computing

With the rapid development of mobile payments, understanding the dynamic relationship between consumer payment behavior and security perception is crucial for enhancing payment security and optimizing user experience. This paper proposes the PaySenseNet model, which …

External Large Foundation Model: How to Efficiently Serve Trillions of Parameters for Online Ads Recommendation

2025 · arXiv (Cornell University)

Ads recommendation is a prominent service of online advertising systems and has been actively studied. Recent studies indicate that scaling-up and advanced design of the recommendation model can bring significant performance improvement. However, with a …

Chinese Medical Question Answer Matching Using End-to-End Character-Level Multi-Scale CNNs

2017 · Applied Sciences

This paper focuses mainly on the problem of Chinese medical question answer matching, which is arguably more challenging than open-domain question answer matching in English due to the combination of its domain-restricted nature and the …