Junshan Zhang
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
Research Interests
Recognitions
Junshan Zhang has a total of 210 co-authors. Here are the top co-authors by number of shared publications:
- Lei Yang
Department of Computer Science and Engineering, University of Nevada, Reno, NV 89557, USA
20 shared publications - Harold Vincent Poor
Princeton University
11 shared publications - Xu Chen
Sun Yat-Sen University, Guangzhou, China
9 shared publications - Xuanyu Cao
School of Electrical Engineering and Computer Science, Washington State University, Pullman, WA, USA
8 shared publications - Xiaowen Gong
Arizona State University
6 shared publications - Xu Chen
University of Göttingen
6 shared publications
Latest Publications
- 15 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Networking
- Vol. 33 (4) , 1870-1885
AugFL: Augmenting Federated Learning With Pretrained Models
Federated Learning (FL) has garnered widespread interest in recent years. However, owing to strict privacy policies or limited storage capacities of training participants such as IoT devices, its effective deployment is often impeded by the scarcity of training data in practical decentralized learning environments. In this paper, we study enhancing FL with the aid of (large) pre-trained models (PMs), that encapsulate wealthy general/domain-agnostic knowledge, to alleviate the data requirement in conducting FL from scratch. Specifically, we consider a networked FL system formed by a central server and distributed clients. First, we formulate the PM-aided personalized FL as a regularization-based federated meta-learning problem, where clients join forces to learn a meta-model with knowledge transferred from a private PM stored at the server. Then, we develop an inexact-ADMM-based algorithm, AugFL, to optimize the problem with no need to expose the PM or incur additional computational costs to local clients. Further, we establish theoretical guarantees for AugFL in terms of communication complexity, adaptation performance, and the benefit of knowledge transfer in general non-convex cases. Extensive experiments corroborate the efficacy and superiority of AugFL over existing baselines.
- 15 April 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Networking
- Vol. 33 (4) , 1870-1885
AugFL: Augmenting Federated Learning With Pretrained Models
Federated Learning (FL) has garnered widespread interest in recent years. However, owing to strict privacy policies or limited storage capacities of training participants such as IoT devices, its effective deployment is often impeded by the scarcity of training data in practical decentralized learning environments. In this paper, we study enhancing FL with the aid of (large) pre-trained models (PMs), that encapsulate wealthy general/domain-agnostic knowledge, to alleviate the data requirement in conducting FL from scratch. Specifically, we consider a networked FL system formed by a central server and distributed clients. First, we formulate the PM-aided personalized FL as a regularization-based federated meta-learning problem, where clients join forces to learn a meta-model with knowledge transferred from a private PM stored at the server. Then, we develop an inexact-ADMM-based algorithm, AugFL, to optimize the problem with no need to expose the PM or incur additional computational costs to local clients. Further, we establish theoretical guarantees for AugFL in terms of communication complexity, adaptation performance, and the benefit of knowledge transfer in general non-convex cases. Extensive experiments corroborate the efficacy and superiority of AugFL over existing baselines.
- 18 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (13) , 22934-22944
EI-Drive: A Platform for Cooperative Perception With Realistic Communication Models
The growing interest in autonomous driving calls for realistic simulation platforms capable of accurately simulating cooperative perception process in realistic traffic scenarios. Existing studies for cooperative perception often have not accounted for transmission latency and errors in real-world environments. To address this gap, we introduce EI-Drive (Edge Intelligent Drive), an Edge-AI based autonomous driving simulation platform that integrates advanced cooperative perception with more realistic communication models. Built on the CARLA framework, EI-Drive features new modules for cooperative perception while taking into account transmission latency and errors, providing a more realistic platform for evaluating cooperative perception algorithms. In particular, the platform enables vehicles to fuse data from multiple sources, improving situational awareness and safety in complex environments. With its modular design, EI-Drive allows for detailed exploration of sensing, perception, planning, and control in various cooperative driving scenarios. Experiments using EI-Drive demonstrate significant improvements in vehicle safety and performance, particularly in scenarios with complex traffic flow and network conditions. All code and documents are accessible on our GitHub page: https://ucd-dare.github.io/eidrive.github.io/.
- 18 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (13) , 22934-22944
EI-Drive: A Platform for Cooperative Perception With Realistic Communication Models
The growing interest in autonomous driving calls for realistic simulation platforms capable of accurately simulating cooperative perception process in realistic traffic scenarios. Existing studies for cooperative perception often have not accounted for transmission latency and errors in real-world environments. To address this gap, we introduce EI-Drive (Edge Intelligent Drive), an Edge-AI based autonomous driving simulation platform that integrates advanced cooperative perception with more realistic communication models. Built on the CARLA framework, EI-Drive features new modules for cooperative perception while taking into account transmission latency and errors, providing a more realistic platform for evaluating cooperative perception algorithms. In particular, the platform enables vehicles to fuse data from multiple sources, improving situational awareness and safety in complex environments. With its modular design, EI-Drive allows for detailed exploration of sensing, perception, planning, and control in various cooperative driving scenarios. Experiments using EI-Drive demonstrate significant improvements in vehicle safety and performance, particularly in scenarios with complex traffic flow and network conditions. All code and documents are accessible on our GitHub page: https://ucd-dare.github.io/eidrive.github.io/.
- 11 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (3) , 2866-2875
CarDreamer: Open-Source Learning Platform for World-Model-Based Autonomous Driving
To safely navigate intricate real-world scenarios, autonomous vehicles (AVs) must be able to adapt to diverse road conditions and anticipate future events. World model (WM)-based reinforcement learning (RL) has emerged as a promising approach by learning and predicting the complex dynamics of various environments. Nevertheless, to the best of our knowledge, there does not exist an open-source platform for training and testing such algorithms in complicated driving environments. To fill this void, we introduce CarDreamer, the first open-source learning platform designed specifically for developing and evaluating WM-based autonomous driving algorithms. It comprises a few key components, including 1) WM Backbone: CarDreamer has integrated some state-of-the-art WMs, which simplifies the reproduction of RL algorithms; 2) Built-In Tasks: CarDreamer offers a comprehensive set of highly configurable driving tasks which are compatible with gym interfaces and are equipped with empirically optimized reward functions; and 3) Task Development Suite: CarDreamer integrates a flexible task development suite to streamline the creation of driving tasks. This suite enables easy definition of traffic flows and vehicle routes, along with automatic collection of multimodal observation data. Furthermore, we conduct extensive experiments using built-in tasks to evaluate the performance and potential of WMs in autonomous driving. Thanks to the richness and flexibility of CarDreamer, we also systematically study the impact of observation modality, observability, and sharing of vehicle intentions on AV safety and efficiency. All code and documents are accessible on our GitHub page https://github.com/ucd-dare/CarDreamer.
- 11 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 12 (3) , 2866-2875
CarDreamer: Open-Source Learning Platform for World-Model-Based Autonomous Driving
To safely navigate intricate real-world scenarios, autonomous vehicles (AVs) must be able to adapt to diverse road conditions and anticipate future events. World model (WM)-based reinforcement learning (RL) has emerged as a promising approach by learning and predicting the complex dynamics of various environments. Nevertheless, to the best of our knowledge, there does not exist an open-source platform for training and testing such algorithms in complicated driving environments. To fill this void, we introduce CarDreamer, the first open-source learning platform designed specifically for developing and evaluating WM-based autonomous driving algorithms. It comprises a few key components, including 1) WM Backbone: CarDreamer has integrated some state-of-the-art WMs, which simplifies the reproduction of RL algorithms; 2) Built-In Tasks: CarDreamer offers a comprehensive set of highly configurable driving tasks which are compatible with gym interfaces and are equipped with empirically optimized reward functions; and 3) Task Development Suite: CarDreamer integrates a flexible task development suite to streamline the creation of driving tasks. This suite enables easy definition of traffic flows and vehicle routes, along with automatic collection of multimodal observation data. Furthermore, we conduct extensive experiments using built-in tasks to evaluate the performance and potential of WMs in autonomous driving. Thanks to the richness and flexibility of CarDreamer, we also systematically study the impact of observation modality, observability, and sharing of vehicle intentions on AV safety and efficiency. All code and documents are accessible on our GitHub page https://github.com/ucd-dare/CarDreamer.
- 16 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 23 (12) , 10952-10965
Impact of Sensing Errors on Headway Design: From -Fair Group Safety to Traffic Throughput
Headway, namely the distance between vehicles, is a key design factor for ensuring the safe operation of autonomous driving systems. There have been studies on headway optimization based on the speeds of leading and trailing vehicles, assuming perfect sensing capabilities. In practical scenarios, however, sensing errors are inevitable, calling for a more robust headway design to mitigate the risk of collision. Undoubtedly, augmenting the safety distance would reduce traffic throughput, thus highlighting the need for headway design to incorporate both sensing errors and risk tolerance models. In addition, prioritizing group safety over individual safety is often deemed unacceptable because no driver should sacrifice their safety for the safety of others. In this study, we propose a multi-objective optimization framework that examines the impact of sensing errors on both traffic throughput and the fairness of safety among vehicles. The proposed framework provides a solution to determine the Pareto frontier for traffic throughput and vehicle safety. ComDrive, a communication-based autonomous driving simulation platform, is developed to validate the proposed approach. Extensive experiments demonstrate that the proposed approach outperforms existing baselines.
- 16 April 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 23 (12) , 10952-10965
Impact of Sensing Errors on Headway Design: From -Fair Group Safety to Traffic Throughput
Headway, namely the distance between vehicles, is a key design factor for ensuring the safe operation of autonomous driving systems. There have been studies on headway optimization based on the speeds of leading and trailing vehicles, assuming perfect sensing capabilities. In practical scenarios, however, sensing errors are inevitable, calling for a more robust headway design to mitigate the risk of collision. Undoubtedly, augmenting the safety distance would reduce traffic throughput, thus highlighting the need for headway design to incorporate both sensing errors and risk tolerance models. In addition, prioritizing group safety over individual safety is often deemed unacceptable because no driver should sacrifice their safety for the safety of others. In this study, we propose a multi-objective optimization framework that examines the impact of sensing errors on both traffic throughput and the fairness of safety among vehicles. The proposed framework provides a solution to determine the Pareto frontier for traffic throughput and vehicle safety. ComDrive, a communication-based autonomous driving simulation platform, is developed to validate the proposed approach. Extensive experiments demonstrate that the proposed approach outperforms existing baselines.
- 18 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 23 (5) , 6104-6115
Towards Resource-Efficient Edge AI: From Federated Learning to Semi-Supervised Model Personalization
A central question in edge intelligence is “how can an edge device learn its local model with limited data and constrained computing capacity?” In this study, we explore the approach where a global model initialization is first obtained by running federated learning (FL) across multiple edge devices, based on which a semi-supervised algorithm is devised for a single edge device to carry out quick adaptation with its local data. Specifically, to account for device heterogeneity and resource constraints, a global model is first trained via FL, where each device conducts multiple local updates only for its customized subnet. A subset of devices can be selected to upload updates for aggregation during each training round. Further, device scheduling is optimized to minimize the training loss of FL, subject to resource constraints, based on the carefully crafted reward function defined as the one-round progress of FL each device can provide. We examine the convergence behavior of FL for the general non-convex case. For semi-supervised model personalization, we use the FL-based model initialization as a teacher network to impute soft labels on unlabeled data, thereby addressing the insufficiency of labeled data. Experiments are conducted to evaluate the performance of the proposed algorithms.
- 18 September 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Mobile Computing
- Vol. 23 (5) , 6104-6115
Towards Resource-Efficient Edge AI: From Federated Learning to Semi-Supervised Model Personalization
A central question in edge intelligence is “how can an edge device learn its local model with limited data and constrained computing capacity?” In this study, we explore the approach where a global model initialization is first obtained by running federated learning (FL) across multiple edge devices, based on which a semi-supervised algorithm is devised for a single edge device to carry out quick adaptation with its local data. Specifically, to account for device heterogeneity and resource constraints, a global model is first trained via FL, where each device conducts multiple local updates only for its customized subnet. A subset of devices can be selected to upload updates for aggregation during each training round. Further, device scheduling is optimized to minimize the training loss of FL, subject to resource constraints, based on the carefully crafted reward function defined as the one-round progress of FL each device can provide. We examine the convergence behavior of FL for the general non-convex case. For semi-supervised model personalization, we use the FL-based model initialization as a teacher network to impute soft labels on unlabeled data, thereby addressing the insufficiency of labeled data. Experiments are conducted to evaluate the performance of the proposed algorithms.
- 14 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 10 (24) , 22331-22340
Scheduling Real-Time Wireless Traffic: A Network-Aided Offline Reinforcement Learning Approach
Real-time traffic has stringent requirements in terms of latency, and deadline guarantees on packet delivery play a vital role in real-time IoT applications. Deadline-aware wireless scheduling of real-time traffic has been a long-standing open problem, despite significant efforts using analytical methods. Departing from the conventional approaches, this work studies deadline-aware traffic scheduling by taking an offline reinforcement learning (RL) approach to train scheduling algorithms, ready to be used for online scheduling. To address the challenges therein, we propose a Network-Aided Offline RL (NA-ORL) framework for deadline-aware scheduling, by making use of the fact that the network dynamics follows a well-defined physics model. Specifically, in NA-ORL the initialization of the scheduling policy is obtained through behavior cloning with a good model-based scheduling algorithm, and the network-aided actor-critic (A-C) method is utilized to train a better scheduling policy with carefully designed states and reward function, thanks to its nature of policy improvement. Building on NA-ORL, we further devise a Network-Aided Offline Meta-RL (NA-MRL) algorithm to deal with the non-stationary network dynamics. Extensive experimental results demonstrate that the proposed NA-ORL and NA-MRL algorithms can achieve better performance over Adaptive Mixing over Non-Dominated links (AMIX-ND) and Largest-Deficit-First (LDF), in various scenarios for the deadline-aware wireless scheduling.
- 14 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Internet of Things Journal
- Vol. 10 (24) , 22331-22340
Scheduling Real-Time Wireless Traffic: A Network-Aided Offline Reinforcement Learning Approach
Real-time traffic has stringent requirements in terms of latency, and deadline guarantees on packet delivery play a vital role in real-time IoT applications. Deadline-aware wireless scheduling of real-time traffic has been a long-standing open problem, despite significant efforts using analytical methods. Departing from the conventional approaches, this work studies deadline-aware traffic scheduling by taking an offline reinforcement learning (RL) approach to train scheduling algorithms, ready to be used for online scheduling. To address the challenges therein, we propose a Network-Aided Offline RL (NA-ORL) framework for deadline-aware scheduling, by making use of the fact that the network dynamics follows a well-defined physics model. Specifically, in NA-ORL the initialization of the scheduling policy is obtained through behavior cloning with a good model-based scheduling algorithm, and the network-aided actor-critic (A-C) method is utilized to train a better scheduling policy with carefully designed states and reward function, thanks to its nature of policy improvement. Building on NA-ORL, we further devise a Network-Aided Offline Meta-RL (NA-MRL) algorithm to deal with the non-stationary network dynamics. Extensive experimental results demonstrate that the proposed NA-ORL and NA-MRL algorithms can achieve better performance over Adaptive Mixing over Non-Dominated links (AMIX-ND) and Largest-Deficit-First (LDF), in various scenarios for the deadline-aware wireless scheduling.
- 26 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Selected Areas in Communications
- Vol. 41 (8) , 2547-2562
Semantic Communications for Wireless Sensing: RIS-Aided Encoding and Self-Supervised Decoding
Semantic communications can reduce the resource consumption by transmitting task-related semantic information extracted from source messages. However, when the source messages are utilized for various tasks, e.g., wireless sensing data for localization and activities detection, semantic communication technique is difficult to be implemented because of the increased processing complexity. In this paper, we propose the inverse semantic communications as a new paradigm. Instead of extracting semantic information from messages, we aim to encode the task-related source messages into a hyper-source message for data transmission or storage. Following this paradigm, we design an inverse semantic-aware wireless sensing framework with three algorithms for data sampling, reconfigurable intelligent surface (RIS)-aided encoding, and self-supervised decoding, respectively. Specifically, on the one hand, we propose a novel RIS hardware design for encoding several signal spectrums into one MetaSpectrum . To select the task-related signal spectrums for achieving efficient encoding, a semantic hash sampling method is introduced. On the other hand, we propose a self-supervised learning method for decoding the MetaSpectrums to obtain the original signal spectrums. Using the sensing data collected from real-world, we show that our framework can reduce the data volume by 95% compared to that before encoding, without affecting the accomplishment of sensing tasks. Moreover, compared with the typically used uniform sampling scheme, the proposed semantic hash sampling scheme can achieve 67% lower mean squared error in recovering the sensing parameters. In addition, experiment results demonstrate that the amplitude response matrix of the RIS enables the encryption of the sensing data.
- 26 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Selected Areas in Communications
- Vol. 41 (8) , 2547-2562
Semantic Communications for Wireless Sensing: RIS-Aided Encoding and Self-Supervised Decoding
Semantic communications can reduce the resource consumption by transmitting task-related semantic information extracted from source messages. However, when the source messages are utilized for various tasks, e.g., wireless sensing data for localization and activities detection, semantic communication technique is difficult to be implemented because of the increased processing complexity. In this paper, we propose the inverse semantic communications as a new paradigm. Instead of extracting semantic information from messages, we aim to encode the task-related source messages into a hyper-source message for data transmission or storage. Following this paradigm, we design an inverse semantic-aware wireless sensing framework with three algorithms for data sampling, reconfigurable intelligent surface (RIS)-aided encoding, and self-supervised decoding, respectively. Specifically, on the one hand, we propose a novel RIS hardware design for encoding several signal spectrums into one MetaSpectrum . To select the task-related signal spectrums for achieving efficient encoding, a semantic hash sampling method is introduced. On the other hand, we propose a self-supervised learning method for decoding the MetaSpectrums to obtain the original signal spectrums. Using the sensing data collected from real-world, we show that our framework can reduce the data volume by 95% compared to that before encoding, without affecting the accomplishment of sensing tasks. Moreover, compared with the typically used uniform sampling scheme, the proposed semantic hash sampling scheme can achieve 67% lower mean squared error in recovering the sensing parameters. In addition, experiment results demonstrate that the amplitude response matrix of the RIS enables the encryption of the sensing data.
- Open Access
Adaptive Ensemble Q-learning: Minimizing Estimation Bias via Error Feedback
The ensemble method is a promising way to mitigate the overestimation issue in Q-learning, where multiple function approximators are used to estimate the action values. It is known that the estimation bias hinges heavily on the ensemble size (i.e., the number of Q-function approximators used in the target), and that determining the `right' ensemble size is highly nontrivial, because of the time-varying nature of the function approximation errors during the learning process. To tackle this challenge, we first derive an upper bound and a lower bound on the estimation bias, based on which the ensemble size is adapted to drive the bias to be nearly zero, thereby coping with the impact of the time-varying approximation errors accordingly. Motivated by the theoretic findings, we advocate that the ensemble method can be combined with Model Identification Adaptive Control (MIAC) for effective ensemble size adaptation. Specifically, we devise Adaptive Ensemble Q-learning (AdaEQ), a generalized ensemble method with two key steps: (a) approximation error characterization which serves as the feedback for flexibly controlling the ensemble size, and (b) ensemble size adaptation tailored towards minimizing the estimation bias. Extensive experiments are carried out to show that AdaEQ can improve the learning performance than the existing methods for the MuJoCo benchmark.
Open AccessAdaptive Ensemble Q-learning: Minimizing Estimation Bias via Error Feedback
The ensemble method is a promising way to mitigate the overestimation issue in Q-learning, where multiple function approximators are used to estimate the action values. It is known that the estimation bias hinges heavily on the ensemble size (i.e., the number of Q-function approximators used in the target), and that determining the `right' ensemble size is highly nontrivial, because of the time-varying nature of the function approximation errors during the learning process. To tackle this challenge, we first derive an upper bound and a lower bound on the estimation bias, based on which the ensemble size is adapted to drive the bias to be nearly zero, thereby coping with the impact of the time-varying approximation errors accordingly. Motivated by the theoretic findings, we advocate that the ensemble method can be combined with Model Identification Adaptive Control (MIAC) for effective ensemble size adaptation. Specifically, we devise Adaptive Ensemble Q-learning (AdaEQ), a generalized ensemble method with two key steps: (a) approximation error characterization which serves as the feedback for flexibly controlling the ensemble size, and (b) ensemble size adaptation tailored towards minimizing the estimation bias. Extensive experiments are carried out to show that AdaEQ can improve the learning performance than the existing methods for the MuJoCo benchmark.
- Top Cited Papers
- 5 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Selected Areas in Communications
- Vol. 41 (7) , 2158-2175
Attention-Aware Resource Allocation and QoE Analysis for Metaverse xURLLC Services
Metaverse encapsulates our expectations of the next-generation Internet, while bringing new key performance indicators (KPIs). Although conventional ultra-reliable and low-latency communications (URLLC) can satisfy objective KPIs, it is difficult to provide a personalized immersive experience that is a distinctive feature of the Metaverse. Since the quality of experience (QoE) can be regarded as a comprehensive KPI, the URLLC is evolved towards the next generation URLLC (xURLLC) with a personalized resource allocation scheme to achieve higher QoE. To deploy Metaverse xURLLC services, we study the interaction between the Metaverse service provider (MSP) and the network infrastructure provider (InP), and provide an optimal contract design framework. Specifically, the utility of the MSP, defined as a function of Metaverse users’ QoE, is to be maximized, while ensuring the incentives of the InP. To model the QoE mathematically, we propose a novel metric named Meta-Immersion that incorporates both the objective KPIs and subjective feelings of Metaverse users. Furthermore, we develop an attention-aware rendering capacity allocation scheme to improve QoE in xURLLC. Using a user-object-attention level dataset, we validate that the xURLLC can achieve an average of 20.1% QoE improvement compared to the conventional URLLC with a uniform resource allocation scheme.
- 5 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal on Selected Areas in Communications
- Vol. 41 (7) , 2158-2175
Top Cited PapersAttention-Aware Resource Allocation and QoE Analysis for Metaverse xURLLC Services
Metaverse encapsulates our expectations of the next-generation Internet, while bringing new key performance indicators (KPIs). Although conventional ultra-reliable and low-latency communications (URLLC) can satisfy objective KPIs, it is difficult to provide a personalized immersive experience that is a distinctive feature of the Metaverse. Since the quality of experience (QoE) can be regarded as a comprehensive KPI, the URLLC is evolved towards the next generation URLLC (xURLLC) with a personalized resource allocation scheme to achieve higher QoE. To deploy Metaverse xURLLC services, we study the interaction between the Metaverse service provider (MSP) and the network infrastructure provider (InP), and provide an optimal contract design framework. Specifically, the utility of the MSP, defined as a function of Metaverse users’ QoE, is to be maximized, while ensuring the incentives of the InP. To model the QoE mathematically, we propose a novel metric named Meta-Immersion that incorporates both the objective KPIs and subjective feelings of Metaverse users. Furthermore, we develop an attention-aware rendering capacity allocation scheme to improve QoE in xURLLC. Using a user-object-attention level dataset, we validate that the xURLLC can achieve an average of 20.1% QoE improvement compared to the conventional URLLC with a uniform resource allocation scheme.
- 15 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 35 (9) , 12042-12056
Continual Learning of Generative Models With Limited Data: From Wasserstein-1 Barycenter to Adaptive Coalescence
Learning generative models is challenging for a network edge node with limited data and computing power. Since tasks in similar environments share a model similarity, it is plausible to leverage pretrained generative models from other edge nodes. Appealing to optimal transport theory tailored toward Wasserstein-1 generative adversarial networks (WGANs), this study aims to develop a framework that systematically optimizes continual learning of generative models using local data at the edge node while exploiting adaptive coalescence of pretrained generative models. Specifically, by treating the knowledge transfer from other nodes as Wasserstein balls centered around their pretrained models, continual learning of generative models is cast as a constrained optimization problem, which is further reduced to a Wasserstein-1 barycenter problem. A two-stage approach is devised accordingly: 1) the barycenters among the pretrained models are computed offline, where displacement interpolation is used as the theoretic foundation for finding adaptive barycenters via a “recursive” WGAN configuration and 2) the barycenter computed offline is used as metamodel initialization for continual learning, and then, fast adaptation is carried out to find the generative model using the local samples at the target edge node. Finally, a weight ternarization method, based on joint optimization of weights and threshold for quantization, is developed to compress the generative model further. Extensive experimental studies corroborate the effectiveness of the proposed framework.
- 15 March 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 35 (9) , 12042-12056
Continual Learning of Generative Models With Limited Data: From Wasserstein-1 Barycenter to Adaptive Coalescence
Learning generative models is challenging for a network edge node with limited data and computing power. Since tasks in similar environments share a model similarity, it is plausible to leverage pretrained generative models from other edge nodes. Appealing to optimal transport theory tailored toward Wasserstein-1 generative adversarial networks (WGANs), this study aims to develop a framework that systematically optimizes continual learning of generative models using local data at the edge node while exploiting adaptive coalescence of pretrained generative models. Specifically, by treating the knowledge transfer from other nodes as Wasserstein balls centered around their pretrained models, continual learning of generative models is cast as a constrained optimization problem, which is further reduced to a Wasserstein-1 barycenter problem. A two-stage approach is devised accordingly: 1) the barycenters among the pretrained models are computed offline, where displacement interpolation is used as the theoretic foundation for finding adaptive barycenters via a “recursive” WGAN configuration and 2) the barycenter computed offline is used as metamodel initialization for continual learning, and then, fast adaptation is carried out to find the generative model using the local samples at the target edge node. Finally, a weight ternarization method, based on joint optimization of weights and threshold for quantization, is developed to compress the generative model further. Extensive experimental studies corroborate the effectiveness of the proposed framework.