检索结果-内蒙古大学图书馆

Federated Adaptation for Foundation Model-based Recommendations

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Zhang, Chunxu Long, Guodong Guo, Hongkuan Fang, Xiao Song, Yang Liu, Zhaojie Zhou, Guorui Zhang, Zijian Liu, Yang Yang, Bo Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education China College of Computer Science and Technology Jilin University China Australian Artificial Intelligence Institute FEIT University of Technology Sydney Australia Kuaishou Technology China Institute for AI Industry Research Tsinghua University China

With the recent success of large language models, particularly foundation models with generalization abilities, applying foundation models for recommendations becomes a new paradigm to improve existing recommendation systems. It becomes a new open challenge to enable the foundation model to capture user preference changes in a timely manner with reasonable communication and computation costs while preserving privacy. This paper proposes a novel federated adaptation mechanism to enhance the foundation model-based recommendation system in a privacy-preserving manner. Specifically, each client will learn a lightweight personalized adapter using its private data. The adapter then collaborates with pre-trained foundation models to provide recommendation service efficiently with fine-grained manners. Importantly, users’ private behavioral data remains secure as it is not shared with the server. This data localization-based privacy preservation is embodied via the federated learning framework. The model can ensure that shared knowledge is incorporated into all adapters while simultaneously preserving each user’s personal preferences. Experimental results on four benchmark datasets demonstrate our method’s superior performance. Implementation code is available to ease reproducibility1 Copyright © 2024, The Authors. All rights reserved.

关键词： Recommender systems

DK-STN: A Domain knowledge Embedded Spatio-Temporal Network Model for MJO Forecast

学校读者我要写书评

暂无评论

SSRN

SSRN 2023年

作者： Li, Hongliang Zhang, Nong Xu, Zhewen Li, Xiang Liu, Changzheng Zhao, Chongbo Wu, Jie College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering Ministry of Education Changchun130012 China China Meteorological Administration Key Laboratory for Climate Prediction Studies National Climate Center Beijing100081 China

Understanding and predicting the Madden-Julian Oscillation (MJO) is fundamental for precipitation forecasting and disaster prevention. To date, long-term and accurate MJO prediction has remained a challenge for researchers. Conventional MJO prediction methods using Numerical Weather Prediction (NWP) are resource-intensive, time-consuming, and highly unstable. While existing Artificial Neural Network (ANN) methods save resources and speed forecasting, their accuracy never reaches the 28 days predicted by the state-of-the-art NWP method, i.e., the ECMWF, since neural networks cannot handle climate data effectively. In this paper, we present a Domain knowledge Embedded Spatio-Temporal Network (DK-STN), a stable neural network model for accurate and efficient MJO forecasting. It combines the benefits of NWP and ANN methods and successfully improves the forecast accuracy of ANN methods while maintaining a high level of efficiency and stability. We begin with a spatial-temporal network (STN) and embed domain knowledge in it using two key methods: (i) applying a domain knowledge enhancement method and (ii) integrating a domain knowledge processing method into network training. We evaluated DK-STN with the 5th generation of ECMWF reanalysis (ERA5) data and compared it with ECMWF. Given 7 days of climate data as input, DK-STN can generate reliable forecasts for the following 28 days in 1-2 seconds, with an error of only 2-3 days in different seasons. DK-STN significantly exceeds ECMWF in that its forecast accuracy is equivalent to ECMWF's, while its efficiency and stability are significantly superior. © 2023, The Authors. All rights reserved.

关键词： Weather forecasting

Low-altitude Friendly-Jamming for Satellite-Maritime Communications via Generative AI-enabled Deep Reinforcement Learning

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Huang, Jiawei Wang, Aimin Sun, Geng Li, Jiahui Wang, Jiacheng Niyato, Dusit Leung, Victor C.M. College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun130012 China College of Computer Science and Technology Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun130012 China College of Computing and Data Science Nanyang Technological University Singapore639798 Singapore Artificial Intelligence Research Institute Shenzhen MSU-BIT University Shenzhen518115 China College of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China Department of Electrical and Computer Engineering The University of British Columbia VancouverV6T 1Z4 Canada

Low Earth Orbit (LEO) satellites can be used to assist maritime wireless communications for data transmission across wide-ranging areas. However, extensive coverage of LEO satellites, combined with openness of channels, can cause the communication process to suffer from security risks. This paper presents a low-altitude friendly-jamming LEO satellite-maritime communication system enabled by a unmanned aerial vehicle (UAV) to ensure data security at the physical layer. Since such a system requires trade-off policies that balance the secrecy rate and energy consumption of the UAV to meet evolving scenario demands, we formulate a secure satellite-maritime communication multi-objective optimization problem (SSMCMOP). In order to solve the dynamic and long-term optimization problem, we reformulate it into a Markov decision process. We then propose a transformer-enhanced soft actor critic (TransSAC) algorithm, which is a generative artificial intelligence-enable deep reinforcement learning approach to solve the reformulated problem, so that capturing global dependencies and diversely exploring weights. Simulation results demonstrate that the TransSAC outperforms various baselines, and achieves an optimal secrecy rate while effectively minimizing the energy consumption of the UAV. Moreover, the results find more suitable constraint values for the system. Copyright © 2025, The Authors. All rights reserved.

关键词： Multiobjective optimization

Privacy-Preserving Multi-Task Online Matching in Mobile Crowdsensing: A Snapshot-Based Approach

学校读者我要写书评

暂无评论

IEEE Internet of Things Journal 2025年

作者： Wei, Kaimin Zhao, Shiting Chen, Jinpeng Pei, Tingrui Sun, Dezhi Jinan University College of Information Science and Technology China Beijing University of Posts and Telecommunications School of Computer Science National Pilot Software Engineering School China Jilin University Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education China Department of Information Science China

With the growing popularity of Mobile Crowdsensing (MCS), online matching has recently attracted considerable attention. However, most previous schemes focused on single-task matching, which limits their practicality in new MCS applications that require multi-task matching. Moreover, most MCS tasks require workers to share locations with the platform, which poses serious privacy concerns. To address this issue, we propose a privacy-preserving multi-task online matching algorithm in a snapshot-based mode (PMS). Specifically, the entire time period is divided into snapshots to reduce the waiting time for newly arrived tasks to be matched. In each snapshot, the planar Laplace-based privacy mechanism is applied to protect worker locations and ensure Ε-geo-indistinguishability. Meanwhile, the Minimum-Cost Maximum-Flow (MCMF)-based multi-task matching mechanism is presented to maximize the task completion rate while minimizing the total travel cost. Experiments on real-world datasets demonstrate that PMS achieves superior task completion, reduced travel costs, and improved privacy preservation compared to existing algorithms. © 2014 IEEE.

关键词： Privacy by design

Nighttime traffic object detection via adaptively integrating event and frame domains

学校读者我要写书评

暂无评论

Fundamental Research 2024年

作者： Jiang, Yu Wang, Yuehang Zhao, Minghao Zhang, Yongji Qi, Hong The College of Computer Science and Technology Jilin University Changchun 130012 China The Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun 130012 China The College of Earth Sciences Jilin University Changchun 130012 China

Intelligent perception is crucial in Intelligent Transportation Systems (ITS), with vision cameras as critical components. However, traditional RGB cameras exhibit a significant decline in performance when capturing nighttime traffic scenes, limiting their effectiveness in supporting ITS. In contrast, event cameras possess a high dynamic range (140 dB vs. 60 dB for traditional cameras), enabling them to overcome frame degradation in low-light conditions. Recently, multimodal learning paradigms have made substantial progress in various vision tasks, such as image-text retrieval. Motivated by this progress, we propose an adaptive selection and fusion detection method that leverages both event and RGB frame domains to optimize nighttime traffic object detection jointly. To address the challenge of unbalanced multimodal data fusion, we design a learnable adaptive selection and fusion module. This module performs feature ranking and fusion in the channel dimension, allowing efficient multimodal fusion. Additionally, we construct a novel multi-level feature pyramid network based on multimodal attention fusion. This network extracts potential features to enhance robustness in detecting nighttime traffic objects. Furthermore, we curate a dataset for nighttime traffic scenarios comprising RGB frames and corresponding event streams. Through experiments, we demonstrate that our proposed method outperforms current state-of-the-art techniques in event-based, frame-based, and event and frame fusion methods. This highlights the effectiveness of integrating the event and frame domains in enhancing nighttime traffic object detection. © 2023 The Authors

关键词： Event camera Intelligent transportation Multimodal learning Nighttime object detection

CCDD: A Tractable Representation for Model Counting and Uniform Sampling

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Lai, Yong Meel, Kuldeep S. Yap, Roland H.C. Key Laboratory of Symbolic Computation and Knowledge Engineering Ministry of Education Jilin University Changchun130012 China School of Computing National University of Singapore Singapore119077 Singapore

knowledge compilation concerns with the compilation of representation languages to target languages supporting a wide range of tractable operations arising from diverse areas of computer science. Tractable target compilation languages are usually achieved by restrictions on the internal nodes (∧ or ∨) of the NNF. In this paper, we propose a new representation language CCDD, which introduces new restrictions on conjunction nodes to capture equivalent literals. We show that CCDD supports two key queries, model counting and uniform samping, in polytime. We present algorithms and a compiler to compile propositional formulas expressed in CNF into CCDD. Experiments over a large set of benchmarks show that our compilation times are better with smaller representations than state-of-art Decision-DNNF, SDD and OBDD[∧] compilers. We apply our techniques to model counting and uniform sampling, and develop model counter and uniform sampler on CNF. Our empirical evaluation demonstrates the following significant improvements: our model counter can solve 885 instances while the prior state of the art solved only 843 instances, representing an improvement of 43 instances;and our uniform sampler can solve 780 instances while the prior state of the art solved only 648 instances, representing an improvement of 132 instances. Copyright © 2022, The Authors. All rights reserved.

关键词： Program compilers

Image retrieval algorithm based on response value center weighted convolution feature

学校读者我要写书评

暂无评论

Jilin Daxue Xuebao (Gongxueban)/Journal of Jilin University (engineering and Technology Edition) 2022年第11期52卷 2669-2675页

作者： Li, Xiao-Ning Zhao, Hong-Wei Zhang, Dan-Yang Zhang, Yuan College of Computer Science and Technology Jilin University Changchun130012 China College of Computer Science and Technology Changchun Normal University Changchun130032 China Key Laboratory of Symbolic Computation and Knowledge Engineering Ministry of Education Jilin University Changchun130012 China

A novel cross-entropy-based loss is proposed，which is applied to a classical convolutional neural network to obtain a better embedding space. A convolutional feature aggregation algorithm based on the weighting of the center of the response value is designed to process the three-dimensional convolutional features obtained by the neural network. The algorithm obtains the spatial weighting coefficient of the feature map by calculating the current position response value and the Gaussian center. The algorithm reduces the loss of information when the 3D convolution feature map is reduced to a 1D image feature descriptor and realizes the enhancement of the target area. Finally，the obtained image feature descriptors are used for retrieval tasks. On the CUB-200-2011 dataset，the effectiveness of the loss function and the response value center weighting algorithm are respectively verified by ablation experiments. Compared with the current retrieval methods，higher precision and recall rates have been obtained on the four datasets of Paris6k，Oxford5k，CUB-200-2011 and CARS196. © 2022 Editorial Board of Jilin University. All rights reserved.

关键词： Convolution

Tc-Sepm: Characterizing Soft Error Resilience of Cnns on Tensor Cores from Program and Microarchitecture Perspectives

学校读者我要写书评

暂无评论

SSRN

SSRN 2023年

作者： Wei, Xiaohui Zhou, Changbao Yue, Hengshan Zhou, Joey Tianyi College of Computer Science and Technology Jilin University Jilin Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering Ministry of Education Jilin University Jilin Changchun130012 China Singapore138632 Singapore Singapore138632 Singapore

As an architectural CNN accelerator integrated into NVIDIA's GPUs, existing research mainly focus on improving the performance of Tensor Core. However, the highly integrated Tensor Cores are vulnerable to transient faults (i.e., soft errors), causing catastrophic consequences in safety-critical applications like automatic driving. Thus, it is imperative to estimate the reliability of CNNs on Tensor Cores. However, it is expensive to obtain a statistically significant resilience profile of CNNs on Tensor Cores with the existing fault injection(FI)-based reliability estimation methods. To this end, we build TC-SEPM to predict the error resilience of CNNs on Tensor Cores instead of FI methods. To ensure the accuracy of TC-SEPM, we first investigate resilience-related features from microarchitecture and program perspectives. Then, leveraging these heuristic features, we train machine learning models to learn the hidden relationship between error resilience and the investigated features, enabling us to predict the impact of soft errors in Tensor Cores on CNN output. Experimental results show that TC-SEPM achieves high accuracy for individual soft error resiliency prediction and overall program resilience estimation while its overhead is only 1/27 of FI methods. Additionally, TC-SEPM can provide valuable insights for programmers or architects to design more robust CNN models on Tensor Cores. © 2023, The Authors. All rights reserved.

关键词： Error correction

Data-Efficient CLIP-Powered Dual-Branch Networks for Source-Free Unsupervised Domain Adaptation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Li, Yongguang Cao, Yueqi Li, Jindong Wang, Qi Wang, Shengsheng College of Software Jilin University Changchun130012 China College of Computer Science and Technology Jilin University Changchun130012 China Key Laboratory of Symbolic Computation and Knowledge Engineering of Ministry of Education Jilin University Changchun130012 China School of Artificial Intelligence Jilin University Changchun130012 China Engineering Research Center of Knowledge-Driven Human-Machine Intelligence Ministry of Education Jilin University Changchun130012 China

Source-free Unsupervised Domain Adaptation (SF-UDA) aims to transfer a model’s performance from a labeled source domain to an unlabeled target domain without direct access to source samples, addressing critical data privacy concerns. However, most existing SF-UDA approaches assume the availability of abundant source domain samples, which is often impractical due to the high cost of data annotation. To address the dual challenges of limited source data and privacy concerns, we introduce a data-efficient, CLIP-powered dual-branch network (CDBN). This architecture consists of a cross-domain feature transfer branch and a target-specific feature learning branch, leveraging high-confidence target domain samples to transfer text features of source domain categories while learning target-specific soft prompts. By fusing the outputs of both branches, our approach not only effectively transfers source domain category semantic information to the target domain but also reduces the negative impacts of noise and domain gaps during target training. Furthermore, we propose an unsupervised optimization strategy driven by accurate classification and diversity, preserving the classification capability learned from the source domain while generating more confident and diverse predictions in the target domain. CDBN achieves near state-of-the-art performance with far fewer source domain samples than existing methods across 31 transfer tasks on seven datasets. © 2024, CC BY.

关键词： Adversarial machine learning