The process of making decisions is something humans do inherently and routinely,to the extent that it appears commonplace. However,in order to achieve good overall performance,decisions must take into account both the...The process of making decisions is something humans do inherently and routinely,to the extent that it appears commonplace. However,in order to achieve good overall performance,decisions must take into account both the outcomes of past decisions and opportunities of future ones. Reinforcement learning,which is fundamental to sequential decision-making,consists of the following components: 1 A set of decisions epochs; 2 A set of environment states; 3 A set of available actions to transition states; 4 State-action dependent immediate rewards for each action.At each decision,the environment state provides the decision maker with a set of available actions from which to choose. As a result of selecting a particular action in the state,the environment generates an immediate reward for the decision maker and shifts to a different state and decision. The ultimate goal for the decision maker is to maximize the total reward after a sequence of time steps.This paper will focus on an archetypal example of reinforcement learning,the stochastic multi-armed bandit problem. After introducing the dilemma,I will briefly cover the most common methods used to solve it,namely the UCB and εn- greedy algorithms. I will also introduce my own greedy implementation,the strict-greedy algorithm,which more tightly follows the greedy pattern in algorithm design,and show that it runs comparably to the two accepted algorithms.展开更多
Artificial intelligence has permeated all aspects of our lives today. However, to make AI behave like real AI, the critical bottleneck lies in the speed of computing. Quantum computers employ the peculiar and unique p...Artificial intelligence has permeated all aspects of our lives today. However, to make AI behave like real AI, the critical bottleneck lies in the speed of computing. Quantum computers employ the peculiar and unique properties of quantum states such as superposition, entanglement, and interference to process information in ways that classical computers cannot. As a new paradigm of computation, quantum computers are capable of performing tasks intractable for classical processors, thus providing a quantum leap in AI research and making the development of real AI a possibility. In this regard, quantum machine learning not only enhances the classical machine learning approach but more importantly it provides an avenue to explore new machine learning models that have no classical counterparts. The qubit-based quantum computers cannot naturally represent the continuous variables commonly used in machine learning, since the measurement outputs of qubit-based circuits are generally discrete. Therefore, a continuous-variable (CV) quantum architecture based on a photonic quantum computing model is selected for our study. In this work, we employ machine learning and optimization to create photonic quantum circuits that can solve the contextual multi-armed bandit problem, a problem in the domain of reinforcement learning, which demonstrates that quantum reinforcement learning algorithms can be learned by a quantum device.展开更多
Federated learning(FL)is an intricate and privacy-preserving technique that enables distributed mobile devices to collaboratively train a machine learning model.However,in real-world FL scenarios,the training performa...Federated learning(FL)is an intricate and privacy-preserving technique that enables distributed mobile devices to collaboratively train a machine learning model.However,in real-world FL scenarios,the training performance is affected by a combination of factors such as the mobility of user devices,limited communication and computational resources,thus making the user scheduling problem crucial.To tackle this problem,we jointly consider the user mobility,communication and computational capacities,and develop a stochastic optimization problem to minimize the convergence time.Specifically,we first establish a convergence bound on the training performance based on the heterogeneity of users’data,and then leverage this bound to derive the participation rate for each user.After deriving the user-specific participation rate,we aim to minimize the training latency by optimizing user scheduling under the constraints of the energy consumption and participation rate.Afterward,we transform this optimization problem to the contextual multi-armed bandit framework based on the Lyapunov method and solve it with the submodular reward enhanced linear upper confidence bound(SR-linUCB)algorithm.Experimental results demonstrate the superiority of our proposed algorithm on the training performance and time consumption compared with stateof-the-art algorithms for both independent and identically distributed(IID)and non-IID settings.展开更多
The cloud platform has limited defense resources to fully protect the edge servers used to process crowd sensing data in Internet of Things.To guarantee the network's overall security,we present a network defense ...The cloud platform has limited defense resources to fully protect the edge servers used to process crowd sensing data in Internet of Things.To guarantee the network's overall security,we present a network defense resource allocation with multi-armed bandits to maximize the network's overall benefit.Firstly,we propose the method for dynamic setting of node defense resource thresholds to obtain the defender(attacker)benefit function of edge servers(nodes)and distribution.Secondly,we design a defense resource sharing mechanism for neighboring nodes to obtain the defense capability of nodes.Subsequently,we use the decomposability and Lipschitz conti-nuity of the defender's total expected utility to reduce the difference between the utility's discrete and continuous arms and analyze the difference theoretically.Finally,experimental results show that the method maximizes the defender's total expected utility and reduces the difference between the discrete and continuous arms of the utility.展开更多
In this paper we consider the contextual multi-armed bandit problem for linear payoffs under a risk-averse criterion.At each round,contexts are revealed for each arm,and the decision maker chooses one arm to pull and ...In this paper we consider the contextual multi-armed bandit problem for linear payoffs under a risk-averse criterion.At each round,contexts are revealed for each arm,and the decision maker chooses one arm to pull and receives the corresponding reward.In particular,we consider mean-variance as the risk criterion,and the best arm is the one with the largest mean-variance reward.We apply the Thompson sampling algorithm for the disjoint model,and provide a comprehensive regret analysis for a variant of the proposed algorithm.For T rounds,K actions,and d-dimensional feature vectors,we prove a regret bound of O((1+ρ+1/ρ)d In T ln K/δ√dKT^(1+2∈)ln K/δ1/e)that holds with probability 1-δunder the mean-variance criterion with risk tolerance p,for any 0<ε<1/2,0<δ<1.The empirical performance of our proposed algorithms is demonstrated via a portfolio selection problem.展开更多
As a combination of edge computing and artificial intelligence,edge intelligence has become a promising technique and provided its users with a series of fast,precise,and customized services.In edge intelligence,when ...As a combination of edge computing and artificial intelligence,edge intelligence has become a promising technique and provided its users with a series of fast,precise,and customized services.In edge intelligence,when learning agents are deployed on the edge side,the data aggregation from the end side to the designated edge devices is an important research topic.Considering the various importance of end devices,this paper studies the weighted data aggregation problem in a single hop end-to-edge communication network.Firstly,to make sure all the end devices with various weights are fairly treated in data aggregation,a distributed end-to-edge cooperative scheme is proposed.Then,to handle the massive contention on the wireless channel caused by end devices,a multi-armed bandit(MAB)algorithm is designed to help the end devices find their most appropriate update rates.Diffe-rent from the traditional data aggregation works,combining the MAB enables our algorithm a higher efficiency in data aggregation.With a theoretical analysis,we show that the efficiency of our algorithm is asymptotically optimal.Comparative experiments with previous works are also conducted to show the strength of our algorithm.展开更多
Mobile Crowdsensing(MCS)represents a transformative approach to collecting data from the environment as it utilizes the ubiquity and sensory capabilities of mobile devices with human participants.This paradigm enables...Mobile Crowdsensing(MCS)represents a transformative approach to collecting data from the environment as it utilizes the ubiquity and sensory capabilities of mobile devices with human participants.This paradigm enables scales of data collection critical for applications ranging from environmental monitoring to urban planning.However,the effective harnessing of this distributed data collection capability faces significant challenges.One of the most significant challenges is the variability in the sensing qualities of the participating devices while they are initially unknown and must be learned over time to optimize task assignments.This paper tackles the dual challenges of managing task diversity to mitigate data redundancy and optimizing task assignment amidst the inherent variability of worker performance.We introduce a novel model that dynamically adjusts task weights based on assignment frequency to promote diversity and incorporates a flexible approach to account for the different qualities of task completion,especially in scenarios with overlapping task assignments.Our strategy aims to maximize the overall weighted quality of data collected within the constraints of a predefined budget.Our strategy leverages a combinatorial multi-armed bandit framework with an upper confidence bound approach to guide decision-making.We demonstrate the efficacy of our approach through a combination of regret analysis and simulations grounded in realistic scenarios.展开更多
In order to solve the high latency of traditional cloud computing and the processing capacity limitation of Internet of Things(IoT)users,Multi-access Edge Computing(MEC)migrates computing and storage capabilities from...In order to solve the high latency of traditional cloud computing and the processing capacity limitation of Internet of Things(IoT)users,Multi-access Edge Computing(MEC)migrates computing and storage capabilities from the remote data center to the edge of network,providing users with computation services quickly and directly.In this paper,we investigate the impact of the randomness caused by the movement of the IoT user on decision-making for offloading,where the connection between the IoT user and the MEC servers is uncertain.This uncertainty would be the main obstacle to assign the task accurately.Consequently,if the assigned task cannot match well with the real connection time,a migration(connection time is not enough to process)would be caused.In order to address the impact of this uncertainty,we formulate the offloading decision as an optimization problem considering the transmission,computation and migration.With the help of Stochastic Programming(SP),we use the posteriori recourse to compensate for inaccurate predictions.Meanwhile,in heterogeneous networks,considering multiple candidate MEC servers could be selected simultaneously due to overlapping,we also introduce the Multi-Arm Bandit(MAB)theory for MEC selection.The extensive simulations validate the improvement and effectiveness of the proposed SP-based Multi-arm bandit Method(SMM)for offloading in terms of reward,cost,energy consumption and delay.The results showthat SMMcan achieve about 20%improvement compared with the traditional offloading method that does not consider the randomness,and it also outperforms the existing SP/MAB based method for offloading.展开更多
The Thresholding Bandit(TB)problem is a popular sequential decision-making problem,which aims at identifying the systems whose means are greater than a threshold.Instead of working on the upper bound of a loss functio...The Thresholding Bandit(TB)problem is a popular sequential decision-making problem,which aims at identifying the systems whose means are greater than a threshold.Instead of working on the upper bound of a loss function,our approach stands out from conventional practices by directly minimizing the loss itself.Leveraging the large deviation theory,we firstly provide an asymptotically optimal allocation rule for the TB problem,and then propose a parameter-free Large Deviation(LD)algorithm to make the allocation rule implementable.Central limit theorem-based Large Deviation(CLD)algorithm is further proposed as a supplement to improve the computation efficiency using normal approximation.Extensive experiments are conducted to validate the superiority of our algorithms compared to existing methods,and demonstrate their broader applications to more general distributions and various kinds of loss functions.展开更多
As the penetration of renewable energy continues to increase,stochastic and intermittent generation resources gradually replace the conventional generators,bringing significant challenges in stabilizing power system f...As the penetration of renewable energy continues to increase,stochastic and intermittent generation resources gradually replace the conventional generators,bringing significant challenges in stabilizing power system frequency.Thus,aggregating demand-side resources for frequency regulation attracts attentions from both academia and industry.However,in practice,conventional aggregation approaches suffer from random and uncertain behaviors of the users such as opting out control signals.The risk-averse multi-armed bandit learning approach is adopted to learn the behaviors of the users and a novel aggregation strategy is developed for residential heating,ventilation,and air conditioning(HVAC)to provide reliable secondary frequency regulation.Compared with the conventional approach,the simulation results show that the risk-averse multiarmed bandit learning approach performs better in secondary frequency regulation with fewer users being selected and opting out of the control.Besides,the proposed approach is more robust to random and changing behaviors of the users.展开更多
In order to cope with the increasing threat of the ballistic missile(BM)in a shorter reaction time,the shooting policy of the layered defense system needs to be optimized.The main decisionmaking problem of shooting op...In order to cope with the increasing threat of the ballistic missile(BM)in a shorter reaction time,the shooting policy of the layered defense system needs to be optimized.The main decisionmaking problem of shooting optimization is how to choose the next BM which needs to be shot according to the previous engagements and results,thus maximizing the expected return of BMs killed or minimizing the cost of BMs penetration.Motivated by this,this study aims to determine an optimal shooting policy for a two-layer missile defense(TLMD)system.This paper considers a scenario in which the TLMD system wishes to shoot at a collection of BMs one at a time,and to maximize the return obtained from BMs killed before the system demise.To provide a policy analysis tool,this paper develops a general model for shooting decision-making,the shooting engagements can be described as a discounted reward Markov decision process.The index shooting policy is a strategy that can effectively balance the shooting returns and the risk that the defense mission fails,and the goal is to maximize the return obtained from BMs killed before the system demise.The numerical results show that the index policy is better than a range of competitors,especially the mean returns and the mean killing BM number.展开更多
文摘The process of making decisions is something humans do inherently and routinely,to the extent that it appears commonplace. However,in order to achieve good overall performance,decisions must take into account both the outcomes of past decisions and opportunities of future ones. Reinforcement learning,which is fundamental to sequential decision-making,consists of the following components: 1 A set of decisions epochs; 2 A set of environment states; 3 A set of available actions to transition states; 4 State-action dependent immediate rewards for each action.At each decision,the environment state provides the decision maker with a set of available actions from which to choose. As a result of selecting a particular action in the state,the environment generates an immediate reward for the decision maker and shifts to a different state and decision. The ultimate goal for the decision maker is to maximize the total reward after a sequence of time steps.This paper will focus on an archetypal example of reinforcement learning,the stochastic multi-armed bandit problem. After introducing the dilemma,I will briefly cover the most common methods used to solve it,namely the UCB and εn- greedy algorithms. I will also introduce my own greedy implementation,the strict-greedy algorithm,which more tightly follows the greedy pattern in algorithm design,and show that it runs comparably to the two accepted algorithms.
文摘Artificial intelligence has permeated all aspects of our lives today. However, to make AI behave like real AI, the critical bottleneck lies in the speed of computing. Quantum computers employ the peculiar and unique properties of quantum states such as superposition, entanglement, and interference to process information in ways that classical computers cannot. As a new paradigm of computation, quantum computers are capable of performing tasks intractable for classical processors, thus providing a quantum leap in AI research and making the development of real AI a possibility. In this regard, quantum machine learning not only enhances the classical machine learning approach but more importantly it provides an avenue to explore new machine learning models that have no classical counterparts. The qubit-based quantum computers cannot naturally represent the continuous variables commonly used in machine learning, since the measurement outputs of qubit-based circuits are generally discrete. Therefore, a continuous-variable (CV) quantum architecture based on a photonic quantum computing model is selected for our study. In this work, we employ machine learning and optimization to create photonic quantum circuits that can solve the contextual multi-armed bandit problem, a problem in the domain of reinforcement learning, which demonstrates that quantum reinforcement learning algorithms can be learned by a quantum device.
文摘Federated learning(FL)is an intricate and privacy-preserving technique that enables distributed mobile devices to collaboratively train a machine learning model.However,in real-world FL scenarios,the training performance is affected by a combination of factors such as the mobility of user devices,limited communication and computational resources,thus making the user scheduling problem crucial.To tackle this problem,we jointly consider the user mobility,communication and computational capacities,and develop a stochastic optimization problem to minimize the convergence time.Specifically,we first establish a convergence bound on the training performance based on the heterogeneity of users’data,and then leverage this bound to derive the participation rate for each user.After deriving the user-specific participation rate,we aim to minimize the training latency by optimizing user scheduling under the constraints of the energy consumption and participation rate.Afterward,we transform this optimization problem to the contextual multi-armed bandit framework based on the Lyapunov method and solve it with the submodular reward enhanced linear upper confidence bound(SR-linUCB)algorithm.Experimental results demonstrate the superiority of our proposed algorithm on the training performance and time consumption compared with stateof-the-art algorithms for both independent and identically distributed(IID)and non-IID settings.
基金supported by the National Natural Science Foundation of China(NSFC)[grant numbers 62172377,61872205]the Shandong Provincial Natural Science Foundation[grant number ZR2019MF018]the Startup Research Foundation for Distinguished Scholars No.202112016.
文摘The cloud platform has limited defense resources to fully protect the edge servers used to process crowd sensing data in Internet of Things.To guarantee the network's overall security,we present a network defense resource allocation with multi-armed bandits to maximize the network's overall benefit.Firstly,we propose the method for dynamic setting of node defense resource thresholds to obtain the defender(attacker)benefit function of edge servers(nodes)and distribution.Secondly,we design a defense resource sharing mechanism for neighboring nodes to obtain the defense capability of nodes.Subsequently,we use the decomposability and Lipschitz conti-nuity of the defender's total expected utility to reduce the difference between the utility's discrete and continuous arms and analyze the difference theoretically.Finally,experimental results show that the method maximizes the defender's total expected utility and reduces the difference between the discrete and continuous arms of the utility.
基金support by the Air Force Office of Scientific Research under Grant FA9550-19-1-0283 and Grant FA9550-22-1-0244National Science Foundation under Grant DMS2053489.
文摘In this paper we consider the contextual multi-armed bandit problem for linear payoffs under a risk-averse criterion.At each round,contexts are revealed for each arm,and the decision maker chooses one arm to pull and receives the corresponding reward.In particular,we consider mean-variance as the risk criterion,and the best arm is the one with the largest mean-variance reward.We apply the Thompson sampling algorithm for the disjoint model,and provide a comprehensive regret analysis for a variant of the proposed algorithm.For T rounds,K actions,and d-dimensional feature vectors,we prove a regret bound of O((1+ρ+1/ρ)d In T ln K/δ√dKT^(1+2∈)ln K/δ1/e)that holds with probability 1-δunder the mean-variance criterion with risk tolerance p,for any 0<ε<1/2,0<δ<1.The empirical performance of our proposed algorithms is demonstrated via a portfolio selection problem.
基金supported by the National Natural Science Foundation of China(NSFC)(62102232,62122042,61971269)Natural Science Foundation of Shandong Province Under(ZR2021QF064)。
文摘As a combination of edge computing and artificial intelligence,edge intelligence has become a promising technique and provided its users with a series of fast,precise,and customized services.In edge intelligence,when learning agents are deployed on the edge side,the data aggregation from the end side to the designated edge devices is an important research topic.Considering the various importance of end devices,this paper studies the weighted data aggregation problem in a single hop end-to-edge communication network.Firstly,to make sure all the end devices with various weights are fairly treated in data aggregation,a distributed end-to-edge cooperative scheme is proposed.Then,to handle the massive contention on the wireless channel caused by end devices,a multi-armed bandit(MAB)algorithm is designed to help the end devices find their most appropriate update rates.Diffe-rent from the traditional data aggregation works,combining the MAB enables our algorithm a higher efficiency in data aggregation.With a theoretical analysis,we show that the efficiency of our algorithm is asymptotically optimal.Comparative experiments with previous works are also conducted to show the strength of our algorithm.
基金supported in part by NSF(Nos.SaTC 2310298,CNS 2214940,CPS 2128378,CNS 2107014,and CNS 2150152).
文摘Mobile Crowdsensing(MCS)represents a transformative approach to collecting data from the environment as it utilizes the ubiquity and sensory capabilities of mobile devices with human participants.This paradigm enables scales of data collection critical for applications ranging from environmental monitoring to urban planning.However,the effective harnessing of this distributed data collection capability faces significant challenges.One of the most significant challenges is the variability in the sensing qualities of the participating devices while they are initially unknown and must be learned over time to optimize task assignments.This paper tackles the dual challenges of managing task diversity to mitigate data redundancy and optimizing task assignment amidst the inherent variability of worker performance.We introduce a novel model that dynamically adjusts task weights based on assignment frequency to promote diversity and incorporates a flexible approach to account for the different qualities of task completion,especially in scenarios with overlapping task assignments.Our strategy aims to maximize the overall weighted quality of data collected within the constraints of a predefined budget.Our strategy leverages a combinatorial multi-armed bandit framework with an upper confidence bound approach to guide decision-making.We demonstrate the efficacy of our approach through a combination of regret analysis and simulations grounded in realistic scenarios.
基金This work was supported in part by the Zhejiang Lab under Grant 20210AB02in part by the Sichuan International Science and Technology Innovation Cooperation/Hong Kong,Macao and Taiwan Science and Technology Innovation Cooperation Project under Grant 2019YFH0163in part by the Key Research and Development Project of Sichuan Provincial Department of Science and Technology under Grant 2018JZ0071.
文摘In order to solve the high latency of traditional cloud computing and the processing capacity limitation of Internet of Things(IoT)users,Multi-access Edge Computing(MEC)migrates computing and storage capabilities from the remote data center to the edge of network,providing users with computation services quickly and directly.In this paper,we investigate the impact of the randomness caused by the movement of the IoT user on decision-making for offloading,where the connection between the IoT user and the MEC servers is uncertain.This uncertainty would be the main obstacle to assign the task accurately.Consequently,if the assigned task cannot match well with the real connection time,a migration(connection time is not enough to process)would be caused.In order to address the impact of this uncertainty,we formulate the offloading decision as an optimization problem considering the transmission,computation and migration.With the help of Stochastic Programming(SP),we use the posteriori recourse to compensate for inaccurate predictions.Meanwhile,in heterogeneous networks,considering multiple candidate MEC servers could be selected simultaneously due to overlapping,we also introduce the Multi-Arm Bandit(MAB)theory for MEC selection.The extensive simulations validate the improvement and effectiveness of the proposed SP-based Multi-arm bandit Method(SMM)for offloading in terms of reward,cost,energy consumption and delay.The results showthat SMMcan achieve about 20%improvement compared with the traditional offloading method that does not consider the randomness,and it also outperforms the existing SP/MAB based method for offloading.
基金supported by the Natural Science Foundation of Guangdong Province(No.2023A1515011667)the Science and Technology Major Project of Shenzhen(No.KJZD20230923114809020)+3 种基金the Key Basic Research Foundation of Shenzhen(No.JCYJ20220818100205012)the Guangdong Basic and Applied Basic Research Foundation(No.2023B1515120020)the Shenzhen Science and Technology Program(No.RCBS20221008093331068)the Hetao Shenzhen-Hong Kong Science and Technology Innovation Cooperation Zone Project(No.HZQSWS-KCCYB-2024016).
文摘The Thresholding Bandit(TB)problem is a popular sequential decision-making problem,which aims at identifying the systems whose means are greater than a threshold.Instead of working on the upper bound of a loss function,our approach stands out from conventional practices by directly minimizing the loss itself.Leveraging the large deviation theory,we firstly provide an asymptotically optimal allocation rule for the TB problem,and then propose a parameter-free Large Deviation(LD)algorithm to make the allocation rule implementable.Central limit theorem-based Large Deviation(CLD)algorithm is further proposed as a supplement to improve the computation efficiency using normal approximation.Extensive experiments are conducted to validate the superiority of our algorithms compared to existing methods,and demonstrate their broader applications to more general distributions and various kinds of loss functions.
基金supported by the National Natural Science Foundation of China(No.51907026)Natural Science Foundation of Jiangsu(No.BK20190361)+1 种基金Jiangsu Provincial Key Laboratory of Smart Grid Technology and EquipmentGlobal Energy Interconnection Research Institute(No.SGGR0000WLJS1900107)
文摘As the penetration of renewable energy continues to increase,stochastic and intermittent generation resources gradually replace the conventional generators,bringing significant challenges in stabilizing power system frequency.Thus,aggregating demand-side resources for frequency regulation attracts attentions from both academia and industry.However,in practice,conventional aggregation approaches suffer from random and uncertain behaviors of the users such as opting out control signals.The risk-averse multi-armed bandit learning approach is adopted to learn the behaviors of the users and a novel aggregation strategy is developed for residential heating,ventilation,and air conditioning(HVAC)to provide reliable secondary frequency regulation.Compared with the conventional approach,the simulation results show that the risk-averse multiarmed bandit learning approach performs better in secondary frequency regulation with fewer users being selected and opting out of the control.Besides,the proposed approach is more robust to random and changing behaviors of the users.
基金supported by the National Natural Science Foundation of China(7170120971771216)+1 种基金Shaanxi Natural Science Foundation(2019JQ-250)China Post-doctoral Fund(2019M653962)
文摘In order to cope with the increasing threat of the ballistic missile(BM)in a shorter reaction time,the shooting policy of the layered defense system needs to be optimized.The main decisionmaking problem of shooting optimization is how to choose the next BM which needs to be shot according to the previous engagements and results,thus maximizing the expected return of BMs killed or minimizing the cost of BMs penetration.Motivated by this,this study aims to determine an optimal shooting policy for a two-layer missile defense(TLMD)system.This paper considers a scenario in which the TLMD system wishes to shoot at a collection of BMs one at a time,and to maximize the return obtained from BMs killed before the system demise.To provide a policy analysis tool,this paper develops a general model for shooting decision-making,the shooting engagements can be described as a discounted reward Markov decision process.The index shooting policy is a strategy that can effectively balance the shooting returns and the risk that the defense mission fails,and the goal is to maximize the return obtained from BMs killed before the system demise.The numerical results show that the index policy is better than a range of competitors,especially the mean returns and the mean killing BM number.
