Develop the foundations for the science of cyber-physical systems (CPS) and intelligent autonom... more Develop the foundations for the science of cyber-physical systems (CPS) and intelligent autonomous systems in smart cities, to assure safety, efficiency and security for application areas such as transportation, healthcare and network systems. Relevant fields: control theory, optimization, machine learning, statistics, game theory, and robotics. Current ongoing research topics: ontology representations for assuring safety of perception, learning and control of connected autonomous vehicles; learning and control based on heterogeneous data in the Internet of Things level; T-SET UTC research on technologies for safe and efficient transportation Research contributions: data-driven real-time robust resource allocation with application in smart cities, especially in ride-sharing of intelligent transportation systems; secure control and attack detection of CPS for DARPA HACMS project; wireless sensor and control networks.
As various smart services are increasingly deployed in modern cities, many unexpected conflicts a... more As various smart services are increasingly deployed in modern cities, many unexpected conflicts arise due to various physical world couplings. Existing solutions for conflict resolution often rely on centralized control to enforce predetermined and fixed priorities of different services, which is challenging due to the inconsistent and private objectives of the services. Also, the centralized solutions miss opportunities to more effectively resolve conflicts according to their spatiotemporal locality of the conflicts. To address this issue, we design a decentralized negotiation and conflict resolution framework named DeResolver, which allows services to resolve conflicts by communicating and negotiating with each other to reach a Pareto-optimal agreement autonomously and efficiently. Our design features a two-step self-supervised learning-based algorithm to predict acceptable proposals and their rankings of each opponent through the negotiation. Our design is evaluated with a smart ...
Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, 2021
Network pruning is a widely used technique to reduce computation cost and model size for deep neu... more Network pruning is a widely used technique to reduce computation cost and model size for deep neural networks. However, the typical three-stage pipeline, i.e., training, pruning and retraining (fine-tuning) significantly increases the overall training trails. In this paper, we develop a systematic weight-pruning optimization approach based on Surrogate Lagrangian relaxation (SLR), which is tailored to overcome difficulties caused by the discrete nature of the weight-pruning problem while ensuring fast convergence. We further accelerate the convergence of the SLR by using quadratic penalties. Model parameters obtained by SLR during the training phase are much closer to their optimal values as compared to those obtained by other state-of-the-art methods. We evaluate the proposed method on image classification tasks using CIFAR-10 and ImageNet, as well as object detection tasks using COCO 2014 and Ultra-Fast-Lane-Detection using TuSimple lane detection dataset. Experimental results dem...
In this paper, we establish a zero-sum, hybrid state stochastic game model for designing defense ... more In this paper, we establish a zero-sum, hybrid state stochastic game model for designing defense policies for cyber-physical systems against different types of attacks. With the increasingly integrated properties of cyber-physical systems (CPS) today, security is a challenge for critical infrastructures. Though resilient control and detecting techniques for a specific model of attack have been proposed, to analyze and design detection and defense mechanisms against multiple types of attacks for CPSs requires new system frameworks. Besides security, other requirements such as optimal control cost also need to be considered. The hybrid game model we propose contains physical states that are described by the system dynamics, and a cyber state that represents the detection mode of the system composed by a set of subsystems. A strategy means selecting a subsystem by combining one controller, one estimator and one detector among a finite set of candidate components at each state. Based on...
2016 ACM/IEEE 7th International Conference on Cyber-Physical Systems (ICCPS), 2016
The objective of the article is to develop a system level control framework, incorporate data inf... more The objective of the article is to develop a system level control framework, incorporate data information with real-time control decisions, balance vacant taxis with minimum total idle driving distance, and consider model uncertainties.
In this paper, we establish a zero-sum, hybrid state stochastic game model for designing defense ... more In this paper, we establish a zero-sum, hybrid state stochastic game model for designing defense policies for cyber-physical systems against different types of attacks. With the increasingly integrated properties of cyber-physical systems (CPS) today, security is a challenge for critical infrastructures. Though resilient control and detecting techniques for a specific model of attack have been proposed, to analyze and design detection and defense mechanisms against multiple types of attacks for CPSs requires new system frameworks. Besides security, other requirements such as optimal control cost also need to be considered. The hybrid game model we propose contains physical states that are described by the system dynamics, and a cyber state that represents the detection mode of the system composed by a set of subsystems. A strategy means selecting a subsystem by combining one controller, one estimator and one detector among a finite set of candidate components at each state. Based on...
Network pruning is a widely used technique to reduce computation cost and model size for deep neu... more Network pruning is a widely used technique to reduce computation cost and model size for deep neural networks. However, the typical three-stage pipeline, i.e., training, pruning and retraining (finetuning) significantly increases the overall training trails. For instance, the retraining process could take up to 80 epochs for ResNet-18 on ImageNet, that is 70% of the original model training trails. In this paper, we develop a systematic weight-pruning optimization approach based on Surrogate Lagrangian relaxation (SLR), which is tailored to overcome difficulties caused by the discrete nature of the weight-pruning problem while ensuring fast convergence. We decompose the weight-pruning problem into subproblems, which are coordinated by updating Lagrangian multipliers. Convergence is then accelerated by using quadratic penalty terms. We evaluate the proposed method on image classification tasks, i.e., ResNet-18 and ResNet-50 using ImageNet, and ResNet-18, ResNet-50 and VGG-16 using CIF...
52nd IEEE Conference on Decision and Control, 2013
ABSTRACT The existing tradeoff between control system performance and the detection rate for repl... more ABSTRACT The existing tradeoff between control system performance and the detection rate for replay attacks highlights the need to provide an optimal control policy that balances the security overhead with control cost. We employ a finite horizon, zero-sum, nonstationary stochastic game approach to minimize the worst-case control and detection cost, and obtain an optimal control policy for switching between control-cost optimal (but nonsecure) and secure (but cost-suboptimal) controllers in presence of replay attacks. To formulate the game, we quantify game parameters using knowledge of the system dynamics, controller design and utilized statistical detector. We show that the optimal strategy for the system exists, and present a suboptimal algorithm used to calculate the system's strategy by combining robust game techniques and a finite horizon stationary stochastic game algorithm. Our approach can be generalized for any system with multiple finite cost, time-invariant linear controllers/estimators/intrusion detectors.
Proceedings of the 4th international conference on Embedded networked sensor systems - SenSys '06, 2006
Extensive empirical studies presented in this paper con- firm that the quality of radio communica... more Extensive empirical studies presented in this paper con- firm that the quality of radio communication between low power sensor devices varies significantly with time and envi- ronment. This phenomenon indicates that the previous topol- ogy control solutions, which use static transmission power, transmission range, and link quality, might not be effective in the physical world. To address this issue, online
Security of cyber-physical systems (CPS) is a challenge for increasingly integrated systems today... more Security of cyber-physical systems (CPS) is a challenge for increasingly integrated systems today. To analyze and design detection and defense mechanisms for CPSs requires new system frameworks. In this paper, we establish a zero-sum hybrid stochastic game model, that can be used for designing defense policies for cyber-physical systems against attackers of different types. The hybrid game model contains physical states described by the system dynamics, and a cyber state that represents the detection mode of the system. A system selects a subsystem by combining one controller, one estimator and one detector among a finite set of candidate components at each state. In order to provide scalable and real-time computation of the switching strategies, we propose a moving-horizon approach to solve the zero-sum hybrid stochastic game, and obtain a saddle-point equilibrium policy for balancing the system's security overhead and control cost. This approach leads to a real-time algorithm ...
Develop the foundations for the science of cyber-physical systems (CPS) and intelligent autonom... more Develop the foundations for the science of cyber-physical systems (CPS) and intelligent autonomous systems in smart cities, to assure safety, efficiency and security for application areas such as transportation, healthcare and network systems. Relevant fields: control theory, optimization, machine learning, statistics, game theory, and robotics. Current ongoing research topics: ontology representations for assuring safety of perception, learning and control of connected autonomous vehicles; learning and control based on heterogeneous data in the Internet of Things level; T-SET UTC research on technologies for safe and efficient transportation Research contributions: data-driven real-time robust resource allocation with application in smart cities, especially in ride-sharing of intelligent transportation systems; secure control and attack detection of CPS for DARPA HACMS project; wireless sensor and control networks.
As various smart services are increasingly deployed in modern cities, many unexpected conflicts a... more As various smart services are increasingly deployed in modern cities, many unexpected conflicts arise due to various physical world couplings. Existing solutions for conflict resolution often rely on centralized control to enforce predetermined and fixed priorities of different services, which is challenging due to the inconsistent and private objectives of the services. Also, the centralized solutions miss opportunities to more effectively resolve conflicts according to their spatiotemporal locality of the conflicts. To address this issue, we design a decentralized negotiation and conflict resolution framework named DeResolver, which allows services to resolve conflicts by communicating and negotiating with each other to reach a Pareto-optimal agreement autonomously and efficiently. Our design features a two-step self-supervised learning-based algorithm to predict acceptable proposals and their rankings of each opponent through the negotiation. Our design is evaluated with a smart ...
Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, 2021
Network pruning is a widely used technique to reduce computation cost and model size for deep neu... more Network pruning is a widely used technique to reduce computation cost and model size for deep neural networks. However, the typical three-stage pipeline, i.e., training, pruning and retraining (fine-tuning) significantly increases the overall training trails. In this paper, we develop a systematic weight-pruning optimization approach based on Surrogate Lagrangian relaxation (SLR), which is tailored to overcome difficulties caused by the discrete nature of the weight-pruning problem while ensuring fast convergence. We further accelerate the convergence of the SLR by using quadratic penalties. Model parameters obtained by SLR during the training phase are much closer to their optimal values as compared to those obtained by other state-of-the-art methods. We evaluate the proposed method on image classification tasks using CIFAR-10 and ImageNet, as well as object detection tasks using COCO 2014 and Ultra-Fast-Lane-Detection using TuSimple lane detection dataset. Experimental results dem...
In this paper, we establish a zero-sum, hybrid state stochastic game model for designing defense ... more In this paper, we establish a zero-sum, hybrid state stochastic game model for designing defense policies for cyber-physical systems against different types of attacks. With the increasingly integrated properties of cyber-physical systems (CPS) today, security is a challenge for critical infrastructures. Though resilient control and detecting techniques for a specific model of attack have been proposed, to analyze and design detection and defense mechanisms against multiple types of attacks for CPSs requires new system frameworks. Besides security, other requirements such as optimal control cost also need to be considered. The hybrid game model we propose contains physical states that are described by the system dynamics, and a cyber state that represents the detection mode of the system composed by a set of subsystems. A strategy means selecting a subsystem by combining one controller, one estimator and one detector among a finite set of candidate components at each state. Based on...
2016 ACM/IEEE 7th International Conference on Cyber-Physical Systems (ICCPS), 2016
The objective of the article is to develop a system level control framework, incorporate data inf... more The objective of the article is to develop a system level control framework, incorporate data information with real-time control decisions, balance vacant taxis with minimum total idle driving distance, and consider model uncertainties.
In this paper, we establish a zero-sum, hybrid state stochastic game model for designing defense ... more In this paper, we establish a zero-sum, hybrid state stochastic game model for designing defense policies for cyber-physical systems against different types of attacks. With the increasingly integrated properties of cyber-physical systems (CPS) today, security is a challenge for critical infrastructures. Though resilient control and detecting techniques for a specific model of attack have been proposed, to analyze and design detection and defense mechanisms against multiple types of attacks for CPSs requires new system frameworks. Besides security, other requirements such as optimal control cost also need to be considered. The hybrid game model we propose contains physical states that are described by the system dynamics, and a cyber state that represents the detection mode of the system composed by a set of subsystems. A strategy means selecting a subsystem by combining one controller, one estimator and one detector among a finite set of candidate components at each state. Based on...
Network pruning is a widely used technique to reduce computation cost and model size for deep neu... more Network pruning is a widely used technique to reduce computation cost and model size for deep neural networks. However, the typical three-stage pipeline, i.e., training, pruning and retraining (finetuning) significantly increases the overall training trails. For instance, the retraining process could take up to 80 epochs for ResNet-18 on ImageNet, that is 70% of the original model training trails. In this paper, we develop a systematic weight-pruning optimization approach based on Surrogate Lagrangian relaxation (SLR), which is tailored to overcome difficulties caused by the discrete nature of the weight-pruning problem while ensuring fast convergence. We decompose the weight-pruning problem into subproblems, which are coordinated by updating Lagrangian multipliers. Convergence is then accelerated by using quadratic penalty terms. We evaluate the proposed method on image classification tasks, i.e., ResNet-18 and ResNet-50 using ImageNet, and ResNet-18, ResNet-50 and VGG-16 using CIF...
52nd IEEE Conference on Decision and Control, 2013
ABSTRACT The existing tradeoff between control system performance and the detection rate for repl... more ABSTRACT The existing tradeoff between control system performance and the detection rate for replay attacks highlights the need to provide an optimal control policy that balances the security overhead with control cost. We employ a finite horizon, zero-sum, nonstationary stochastic game approach to minimize the worst-case control and detection cost, and obtain an optimal control policy for switching between control-cost optimal (but nonsecure) and secure (but cost-suboptimal) controllers in presence of replay attacks. To formulate the game, we quantify game parameters using knowledge of the system dynamics, controller design and utilized statistical detector. We show that the optimal strategy for the system exists, and present a suboptimal algorithm used to calculate the system's strategy by combining robust game techniques and a finite horizon stationary stochastic game algorithm. Our approach can be generalized for any system with multiple finite cost, time-invariant linear controllers/estimators/intrusion detectors.
Proceedings of the 4th international conference on Embedded networked sensor systems - SenSys '06, 2006
Extensive empirical studies presented in this paper con- firm that the quality of radio communica... more Extensive empirical studies presented in this paper con- firm that the quality of radio communication between low power sensor devices varies significantly with time and envi- ronment. This phenomenon indicates that the previous topol- ogy control solutions, which use static transmission power, transmission range, and link quality, might not be effective in the physical world. To address this issue, online
Security of cyber-physical systems (CPS) is a challenge for increasingly integrated systems today... more Security of cyber-physical systems (CPS) is a challenge for increasingly integrated systems today. To analyze and design detection and defense mechanisms for CPSs requires new system frameworks. In this paper, we establish a zero-sum hybrid stochastic game model, that can be used for designing defense policies for cyber-physical systems against attackers of different types. The hybrid game model contains physical states described by the system dynamics, and a cyber state that represents the detection mode of the system. A system selects a subsystem by combining one controller, one estimator and one detector among a finite set of candidate components at each state. In order to provide scalable and real-time computation of the switching strategies, we propose a moving-horizon approach to solve the zero-sum hybrid stochastic game, and obtain a saddle-point equilibrium policy for balancing the system's security overhead and control cost. This approach leads to a real-time algorithm ...
Uploads
Papers by Fei Miao