IEEE/ACM International Symposium on Quality of Service (IWQoS) 2022
IEEE/ACM IWQoS Opening
Session Chair
Yan Zhang, University of Oslo, Norway
Keynote 1: DeepFake Detection: Applicability and Robustness
DeepFake Detection: Applicability and Robustness
Kui Ren, IEEE/ACM Fellow, Zhejiang University, China
Session Chair
Yan Zhang, University of Oslo, Norway
Best Paper and IWQoS 2023
Session Chair
Yan Zhang, University of Oslo, Norway
QoS in AI/ML
AIQoSer: Building the efficient Inference-QoS for AI Services
Jianxin Li, Tianchen Zhu, Haoyi Zhou, Qingyun Sun, Chunyang Jiang, Shuai Zhang and Chunming Hu (Beihang University, China)
However, the increasing demands of satisfactory experiences require larger AI models, whose inference efficiency becomes the non-negligible drawback in the time-sensitive network QoS. In this work, we defined this challenge as the inference-QoS (iQoS) problem of the network QoS itself, which balances inference efficiency and performance for AI services. We design a unified iQoS metric to evaluate the AI-enhanced QoS frameworks with considerations on model performance, inference latency, and input scale. Then, we propose a two-stage pipeline as the exemplar for leveraging the iQoS metric in QoS-aware AI services: (i) enhance reconstruction ability, pretraining masked autoencoder extracts intrinsic data correlations by multi-scale masking; (ii) improve inference efficiency, forecasting masked decoder uses the data scale pruning in terms of spatial and temporal dimension for prediction. Comprehensive experiments on our method demonstrate its superior inference latency and overwhelming traffic matrix prediction performance.
Accelerating Blockchain-enabled Distributed Machine Learning by Proof of Useful Work
Yao Du and Cyril Leung (The University of British Columbia, Canada); Zehua Wang (The University of British Columbia, Vancouver, Canada); Victor C.M. Leung (Shenzhen University, China & The University of British Columbia, Canada)
An Online Approach for DNN Model Caching and Processor Allocation in Edge Computing
Zhiqi Chen, Sheng Zhang, Zhi Ma, Shuai Zhang and Zhuzhong Qian (Nanjing University, China); Mingjun Xiao (University of Science and Technology of China, China); Jie Wu (Temple University, USA); Sanglu Lu (Nanjing University, China)
PRM: An Efficient Partial Recovery Method to Accelerate Training Data Reconstruction for Distributed Deep Learning Applications in Cloud Storage Systems
Piao Hu, Yunfei Gu, Ranhao Jia, Chentao Wu and Minyi Guo (Shanghai Jiao Tong University, China); Jie Li (Shanghai Jiaotong University, China)
Quality-aided Annotation Service Selection in MLaaS Market
Shanyang Jiang and Lan Zhang (University of Science and Technology of China, China)
Session Chair
Celimuge Wu, The University of Electro-Communications, Japan
Keynote 2: Quality of Service in the Coming Quantum Internet
Quality of Service in the Coming Quantum Internet
Donald F. Towsley, IEEE/ACM Fellow, University of Massachusetts, Amherst
Session Chair
Ahmed Elmokashfi, SimulaMet, Norway
Federated Learning
How Asynchronous can Federated Learning Be?
Ningxin Su and Baochun Li (University of Toronto, Canada)
In this paper, we seek to explore the entire design space between fully synchronous and asynchronous mechanisms of communication. Based on insights from our exploration, we pro- pose PORT, a new partially asynchronous mechanism designed to allow fast clients to aggregate asynchronously, yet without waiting excessively for the slower ones. In addition, PORT is designed to adjust the aggregation weights based on both the staleness and divergence of model updates, with provable convergence guarantees. We have implemented PORT and its leading competitors in ANONYMOUS, an open-source scalable federated learning research framework designed from the ground up to emulate real-world scenarios. With respect to the wall-clock time it takes for converging to the target accuracy, PORT outperformed leading competitors, including FedBuff and FedAsync, by at least 31% in our experiments.
Federated Graph Learning with Periodic Neighbour Sampling
Bingqian Du and Chuan Wu (The University of Hong Kong, Hong Kong)
Adaptive Clustered Federated Learning for Clients with Time-Varying Interests
Ne Wang, Ruiting Zhou, Lina Su and Guang Fang (Wuhan University, China); Zongpeng Li (Tsinghua University, China)
FedSyL: Computation-Efficient Federated Synergy Learning on heterogeneous IoT devices
Hui Jiang (Institute of Computing Technology, Chinese Academy of Sciences & University of Chinese Academy of Sciences, China); Min Liu and Sheng Sun (Institute of Computing Technology, Chinese Academy of Sciences, China); Yuwei Wang (Institute of Computing Technology Chinese Academy of Sciences, China); Xiaobing Guo (Lenovo Corporate Research & Development, China)
Session Chair
Baochun Li, University of Toronto, Canada
Inter-Networking Communications
Towards Sustainable Multi-Tier Space Networking for LEO Satellite Constellations
Yi Ching Chou (Simon Fraser University, Canada); Xiaoqiang Ma (Douglas College, Canada); Feng Wang (University of Mississippi, USA); Sami Ma (Simon Fraser University, Canada); Sen Hung Wong (Hong Kong University of Science and Technology, Hong Kong); Jiangchuan Liu (Simon Fraser University, Canada)
DoCile: Taming Denial-of-Capability Attacks in Inter-Domain Communications
Marc Wyss, Giacomo Giuliari and Markus Legner (ETH Zürich, Switzerland); Adrian Perrig (ETH Zürich Switzerland & Carnegie Mellon University, USA)
Geographic Low-Earth-Orbit Networking without QoS Bottlenecks from Infrastructure Mobility
Lixin Liu, Hewu Li, Yuanjie Li, Zeqi Lai, Yangtao Deng and Yimei Chen (Tsinghua University, China); Wei Liu (Tsinghua, China); Qian Wu (Tsinghua University, China)
FABMon: Enabling Fast and Accurate Network Available Bandwidth Estimation
Tao Jin (Tsinghua Shenzhen International Graduate School, China); Weichao Li and Qing Li (Peng Cheng Laboratory, China); Qianyi Huang (Southern University of Science and Technology & Peng Cheng Laboratory, China); Yong Jiang (Graduate School at Shenzhen, Tsinghua University, China); Shutao Xia (Tsinghua University, China)
BQR first induces an instant network congestion and then observes the one-way delay (OWD) variation until the tight link recovers from the congestion. By correlating the OWDs with the queue length variation, BQR can calculate the ABW accurately. Compared to the traditional probe gap model (PGM) and probe rate model (PRM), our theoretical analysis and simulations show that BQR is more tolerant to the transient traffic burst and supports the scenarios with multiple congestible links. Based on the model, we build FABMon, a fast and accurate ABW estimation tool. Our experiments show that FABMon can measure ABW within 50 milliseconds, and achieve much more accurate measurement results than the existing tools with a very small volume of probe packets.
Session Chair
Foivos Ioannis Michelinakis, SimulaMet, Norway
Congestion Control
A Control-Theoretic and Online Learning Approach to Self-Tuning Queue Management
Jiancheng Ye (Huawei, Hong Kong); Kechao Cai (Sun Yat-Sen University, China); Dong Lin (Huawei, Hong Kong); Jiarong Li (Tsinghua University, China); Jianfei He (City University of Hong Kong, Hong Kong); John C.S. Lui (The Chinese University of Hong Kong, Hong Kong)
When Power-of-d-Choices Meets Priority
Jianyu Niu (Southern University of Science and Technology, China); Chunpu Wang (Canada); Chen Feng (University of British Columbia, Canada); Hong Xu (The Chinese University of Hong Kong, Hong Kong)
On the Joint Optimization of Function Assignment and Communication Scheduling toward Performance Efficient Serverless Edge Computing
Yuepeng Li and Deze Zeng (China University of Geosciences, China); Lin Gu (Huazhong University of Science and Technology, China); Kun Wang (University of California, Los Angeles, USA); Song Guo (The Hong Kong Polytechnic University, Hong Kong)
Congestion-Aware Modeling and Analysis of Sponsored Data Plan from End User Perspective
Yi Zhao and Qi Tan (Tsinghua University, China); Xiaohua Xu (University of Science and Technology of China, China); Hui Su (Tsinghua University, China); Dan Wang (The Hong Kong Polytechnic University, Hong Kong); Ke Xu (Tsinghua University, China)
Session Chair
Deze Zeng, China University of Geosciences
Made with in Toronto · IWQoS 2020 · IWQoS 2021 · Privacy Policy · © 2022 Duetone Corp.