Chen Yang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Chen Yang has a total of 34 co-authors. Here are the top co-authors by number of shared publications:
- Mengwei Xu
Beijing University of Posts and Telecommunications
5 shared publications - Shangguang Wang
Beijing University of Posts and Telecommunications
5 shared publications - Fei Teng
Southwest Jiaotong University
3 shared publications - Jinliang Yuan
Tsinghua University, Beijing, China
3 shared publications - Qibo Sun
State Key Laboratory of Networking and Switching Technology, Beijing University of Posts and Telecommunications, Beijing, China
3 shared publications - Qiyang Zhang
Peking University
3 shared publications
Latest Publications
- 1 December 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25 (12) , 22835-22848
Communication-Efficient Satellite-Ground Federated Learning Over Intermittent LEO Satellite Networks
- 1 December 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25 (12) , 22835-22848
Communication-Efficient Satellite-Ground Federated Learning Over Intermittent LEO Satellite Networks
- 1 December 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25 (12) , 22520-22535
FeS: Federated NLP in Few-Shot Scenarios
- 1 December 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 25 (12) , 22520-22535
FeS: Federated NLP in Few-Shot Scenarios
- Open Access
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 5488-5499
Taming Multi-Dimensional Tail: A Factorized Tail-Interaction Framework for Hyperscale Cloud Workload Forecasting
The adoption of AI-native cloud infrastructure is reshaping workload behavior in large-scale data centers, moving from stationary, homogeneous patterns to highly dynamic, heterogeneous ones. An empirical analysis of production cloud monitoring data shows that system reliability is increasingly governed by non-linear interactions in the multi-dimensional tail space, including long-range temporal backlogs, spatial resource fragmentation, and irreversible state rigidity. Accurately forecasting these tail-dominated behaviors is essential for capacity planning and reliability management. However, existing forecasting models face two limitations: learning is dominated by abundant normal workloads, leading to rare but critical tail events being underrepresented, and reliance on linear composition assumptions limits the ability to model joint tail anomalies across correlated temporal, spatial, and state dimensions. To address these challenges, we introduce the F actorized T ail- I nteraction F ramework (FTIF), a data-driven forecasting and risk perception framework explicitly designed for tail-sensitive cloud workloads. Firstly, an adaptive trend decoupler separates dominant workload trends from tail residuals using a gradient detachment mechanism and learnable gating mechanisms. Secondly, a factorized resonance module augments attention mechanisms with hyper-order factorization, injecting multiplicative inductive biases to capture non-linear interactions among multi-dimensional risk factors. Thirdly, a rank-rectified tail-focal optimization employs topology-preserving reweighting to align the learning objective with rare but operationally critical tail events. Experiments on real-world cloud datasets show that FTIF achieves an MSE of 0.0015 on the GenTD26 dataset and captures 94.95% of tail events in Spot scenarios, enabling finer-grained risk perception and robust modeling of joint tail anomalies.
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 5488-5499
Open AccessTaming Multi-Dimensional Tail: A Factorized Tail-Interaction Framework for Hyperscale Cloud Workload Forecasting
The adoption of AI-native cloud infrastructure is reshaping workload behavior in large-scale data centers, moving from stationary, homogeneous patterns to highly dynamic, heterogeneous ones. An empirical analysis of production cloud monitoring data shows that system reliability is increasingly governed by non-linear interactions in the multi-dimensional tail space, including long-range temporal backlogs, spatial resource fragmentation, and irreversible state rigidity. Accurately forecasting these tail-dominated behaviors is essential for capacity planning and reliability management. However, existing forecasting models face two limitations: learning is dominated by abundant normal workloads, leading to rare but critical tail events being underrepresented, and reliance on linear composition assumptions limits the ability to model joint tail anomalies across correlated temporal, spatial, and state dimensions. To address these challenges, we introduce the F actorized T ail- I nteraction F ramework (FTIF), a data-driven forecasting and risk perception framework explicitly designed for tail-sensitive cloud workloads. Firstly, an adaptive trend decoupler separates dominant workload trends from tail residuals using a gradient detachment mechanism and learnable gating mechanisms. Secondly, a factorized resonance module augments attention mechanisms with hyper-order factorization, injecting multiplicative inductive biases to capture non-linear interactions among multi-dimensional risk factors. Thirdly, a rank-rectified tail-focal optimization employs topology-preserving reweighting to align the learning objective with rare but operationally critical tail events. Experiments on real-world cloud datasets show that FTIF achieves an MSE of 0.0015 on the GenTD26 dataset and captures 94.95% of tail events in Spot scenarios, enabling finer-grained risk perception and robust modeling of joint tail anomalies.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Services Computing
- Vol. 19 (4) , 1-14
A Provably Cost-Efficient Approach to Deploying MoE Inference Models at the Network Edge
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Services Computing
- Vol. 19 (4) , 1-14
A Provably Cost-Efficient Approach to Deploying MoE Inference Models at the Network Edge
- 25 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Services Computing
- Vol. 18 (4) , 2075-2088
FedCLR+: Tackling Onboard Label Constraints for Accurate Federated Satellite Computing
The rapid growth of Low Earth Orbit (LEO) satellites, particularly with the increasing deployment of intelligent computing capabilities using commercial off-the-shelf (COTS) hardware, presents significant opportunities to enhance the quality of in-orbit services. However, the current onboard conditions remain insufficient to enhance model accuracy by increasing model size, and inadequate accuracy hampers the effectiveness of in-orbit services. The satellite-ground federated learning (FL) paradigm, leveraging collaborative fine-tuning, offers a promising solution to continuously improve onboard model performance. Prior studies have focused on optimizing fine-tuning under constraints like limited bandwidth and computational resources, they often overlook two critical challenges: the scarcity and skewness of labeled onboard data and the long revisit cycles of satellites. To address these challenges and better support in-orbit services, this paper designs a realistic simulation methodology for the onboard fine-tuning process and conducts a comprehensive measurement study. Based on insights from the measurement results, we propose an efficient satellite-ground federated fine-tuning system, FedCLR+. In this system, we design a FedCLR algorithm to enhance system accuracy through representation optimization. Additionally, we propose a hybrid bias-compensated strategy to further mitigate accuracy loss by enriching the diversity of aggregation information. Experimental results show that FedCLR+ significantly enhances accuracy by up to 21.61×, reduces transmission volume by an average of 7.29%, and maintaining acceptable additional overhead compared to baselines.
- 25 June 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Services Computing
- Vol. 18 (4) , 2075-2088
FedCLR+: Tackling Onboard Label Constraints for Accurate Federated Satellite Computing
The rapid growth of Low Earth Orbit (LEO) satellites, particularly with the increasing deployment of intelligent computing capabilities using commercial off-the-shelf (COTS) hardware, presents significant opportunities to enhance the quality of in-orbit services. However, the current onboard conditions remain insufficient to enhance model accuracy by increasing model size, and inadequate accuracy hampers the effectiveness of in-orbit services. The satellite-ground federated learning (FL) paradigm, leveraging collaborative fine-tuning, offers a promising solution to continuously improve onboard model performance. Prior studies have focused on optimizing fine-tuning under constraints like limited bandwidth and computational resources, they often overlook two critical challenges: the scarcity and skewness of labeled onboard data and the long revisit cycles of satellites. To address these challenges and better support in-orbit services, this paper designs a realistic simulation methodology for the onboard fine-tuning process and conducts a comprehensive measurement study. Based on insights from the measurement results, we propose an efficient satellite-ground federated fine-tuning system, FedCLR+. In this system, we design a FedCLR algorithm to enhance system accuracy through representation optimization. Additionally, we propose a hybrid bias-compensated strategy to further mitigate accuracy loss by enriching the diversity of aggregation information. Experimental results show that FedCLR+ significantly enhances accuracy by up to 21.61×, reduces transmission volume by an average of 7.29%, and maintaining acceptable additional overhead compared to baselines.
- 30 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Services Computing
- Vol. 17 (6) , 4411-4422
Toward Efficient Satellite Computing Through Adaptive Compression
The rapid development of Low Earth Orbit (LEO) satellite constellations offers significant potential for in-orbit services, particularly in mitigating the impact of sudden natural disasters. However, the massive data collected by these satellites are often large and severely constrained by limited transmission capabilities when sending data to the ground. Satellite computing, which utilizes onboard computational capacity to process data before transmission, presents a promising solution to alleviate the downlink burden. Nonetheless, this paradigm introduces another bottleneck: limited onboard computing capacity, resulting in slow in-orbit processing and poor results. Current satellite computing systems struggle to efficiently address both data transmission and computing bottlenecks, particularly for urgent disaster services that demand accurate and timely results. Thus, we introduce an efficient satellite computing system designed to jointly mitigate these bottlenecks, thereby providing better service. The core idea is to utilize onboard computing capacity for swift in-orbit annotation of image regions, enabling adaptive compression and download based on annotation confidence and perceived downlink availability. Once the data is downloaded, image restoration and re-inference are performed on the ground to enhance accuracy. Compared to satellite-only inference, our system demonstrates an average improvement in inference accuracy of 3.8%. Furthermore, compared to ground-only inference, with only a 2.8% accuracy loss, our system achieves a 38.4% reduction in response time and saves 71.6% of downlink volume on average.
- 30 September 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Services Computing
- Vol. 17 (6) , 4411-4422
Toward Efficient Satellite Computing Through Adaptive Compression
The rapid development of Low Earth Orbit (LEO) satellite constellations offers significant potential for in-orbit services, particularly in mitigating the impact of sudden natural disasters. However, the massive data collected by these satellites are often large and severely constrained by limited transmission capabilities when sending data to the ground. Satellite computing, which utilizes onboard computational capacity to process data before transmission, presents a promising solution to alleviate the downlink burden. Nonetheless, this paradigm introduces another bottleneck: limited onboard computing capacity, resulting in slow in-orbit processing and poor results. Current satellite computing systems struggle to efficiently address both data transmission and computing bottlenecks, particularly for urgent disaster services that demand accurate and timely results. Thus, we introduce an efficient satellite computing system designed to jointly mitigate these bottlenecks, thereby providing better service. The core idea is to utilize onboard computing capacity for swift in-orbit annotation of image regions, enabling adaptive compression and download based on annotation confidence and perceived downlink availability. Once the data is downloaded, image restoration and re-inference are performed on the ground to enhance accuracy. Compared to satellite-only inference, our system demonstrates an average improvement in inference accuracy of 3.8%. Furthermore, compared to ground-only inference, with only a 2.8% accuracy loss, our system achieves a 38.4% reduction in response time and saves 71.6% of downlink volume on average.
- Open Access
- 29 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 279-295
Mobile Foundation Model as Firmware
- 29 May 2024
- Published by Association for Computing Machinery (ACM)
- p. 279-295
Open AccessMobile Foundation Model as Firmware
- 26 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 23 (9) , 8999-9011
Communication-Efficient Satellite-Ground Federated Learning Through Progressive Weight Quantization
Large constellations of Low Earth Orbit (LEO) satellites have been launched for Earth observation and satellite-ground communication, which collect massive imagery and sensor data. These data can enhance the AI capabilities of satellites to address global challenges such as real-time disaster navigation and mitigation. Prior studies proposed leveraging federated learning (FL) across satellite-ground to collaboratively train a share machine learning (ML) model in a privacy-preserving mechanism. However, they mostly focus on single unique challenges such as limited ground-to-satellite bandwidth, short connection window, and long connection cycle, while ignoring the completeness of these challenges in deploying efficient FL frameworks in space. In this paper, we propose an efficient satellite-ground FL framework, SatelliteFL, to address these three challenges collectively. Its key idea is to ensure that each satellite must complete per-round training within each connection window. Moreover, we design a progressive block-wise quantization algorithm that determines a unique bitwidth for each block of the ML model to maximize the model utility while not exceeding the connection window. We evaluate SatelliteFL by plugging an implemented FL platform into real-world satellite networks and satellite images. The results show that SatelliteFL highly accelerates the convergence by up to 2.8× and improves the bandwidth utilization ratio by up to 9.3× compared to the state-of-the-art methods.
- 26 January 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 23 (9) , 8999-9011
Communication-Efficient Satellite-Ground Federated Learning Through Progressive Weight Quantization
Large constellations of Low Earth Orbit (LEO) satellites have been launched for Earth observation and satellite-ground communication, which collect massive imagery and sensor data. These data can enhance the AI capabilities of satellites to address global challenges such as real-time disaster navigation and mitigation. Prior studies proposed leveraging federated learning (FL) across satellite-ground to collaboratively train a share machine learning (ML) model in a privacy-preserving mechanism. However, they mostly focus on single unique challenges such as limited ground-to-satellite bandwidth, short connection window, and long connection cycle, while ignoring the completeness of these challenges in deploying efficient FL frameworks in space. In this paper, we propose an efficient satellite-ground FL framework, SatelliteFL, to address these three challenges collectively. Its key idea is to ensure that each satellite must complete per-round training within each connection window. Moreover, we design a progressive block-wise quantization algorithm that determines a unique bitwidth for each block of the ML model to maximize the model utility while not exceeding the connection window. We evaluate SatelliteFL by plugging an implemented FL platform into real-world satellite networks and satellite images. The results show that SatelliteFL highly accelerates the convergence by up to 2.8× and improves the bandwidth utilization ratio by up to 9.3× compared to the state-of-the-art methods.