arXiv subjects

Mingyan Liu

Publications and source records attributed to Mingyan Liu.

At least 73 records · Page 4Linked to original sources

arXiv2014

Closing the Price of Anarchy Gap in the Interdependent Security Game

The reliability and security of a user in an interconnected system depends on all users' collective effort in security. Consequently, investments in security technologies by strategic users is typically modeled as a public good problem, known as the Interdependent Security (IDS) game. The equilibria for such games are often inefficient, as selfish users free-ride on positive externalities of others' contributions. In this paper, we present a mechanism that implements the socially optimal equilibrium in an IDS game through a message exchange process, in which users submit proposals about the security investment and tax/price profiles of one another. This mechanism is different from existing solutions in that (1) it results in socially optimal levels of investment, closing the Price of Anarchy gap in the IDS game, (2) it is applicable to a general model of user interdependencies. We further consider the issue of individual rationality, often a trivial condition to satisfy in many resource allocation problems, and argue that with positive externality, the incentive to stay out and free-ride on others' investment can make individual rationality much harder to satisfy in designing a mechanism.

cs.GT↗

arXiv2013

Revisiting Optimal Power Control: its Dual Effect on SNR and Contention

In this paper we study a transmission power tune problem with densely deployed 802.11 Wireless Local Area Networks (WLANs). While previous papers emphasize on tuning transmission power with either PHY or MAC layer separately, optimally setting each Access Point's (AP's) transmission power of a densely deployed 802.11 network considering its dual effects on both layers remains an open problem. In this work, we design a measure by evaluating impacts of transmission power on network performance on both PHY and MAC layers. We show that such an optimization problem is intractable and then we investigate and develop an analytical framework to allow simple yet efficient solutions. Through simulations and numerical results, we observe clear benefits of the dual-effect model compared to solutions optimizing solely on a single layer; therefore, we conclude that tuning transmission power from a dual layer (PHY-MAC) point of view is essential and necessary for dense WLANs. We further form a game theoretical framework and investigate above power-tune problem in a strategic network. We show that with decentralized and strategic users, the Nash Equilibrium (N.E.) of the corresponding game is in-efficient and thereafter we propose a punishment based mechanism to enforce users to adopt the social optimal strategy profile under both perfect and imperfect sensing environments.

eess.SY↗

arXiv2013

Sufficient Conditions on the Optimality of Myopic Sensing in Opportunistic Channel Access: A Unifying Framework

This paper considers a widely studied stochastic control problem arising from opportunistic spectrum access (OSA) in a multi-channel system, with the goal of providing a unifying analytical framework whereby a number of prior results may be viewed as special cases. Specifically, we consider a single wireless transceiver/user with access to $N$ channels, each modeled as an iid discrete-time two-state Markov chain. In each time step the user is allowed to sense $k\leq N$ channels, and subsequently use up to $m\leq k$ channels out of those sensed to be available. Channel sensing is assumed to be perfect, and for each channel use in each time step the user gets a unit reward. The user's objective is to maximize its total discounted or average reward over a finite or infinite horizon. This problem has previously been studied in various special cases including $k=1$ and $m=k\leq N$, often cast as a restless bandit problem, with optimality results derived for a myopic policy that seeks to maximize the immediate one-step reward when the two-state Markov chain model is positively correlated. In this paper we study the general problem with $1\leq m\leq k\leq N$, and derive sufficient conditions under which the myopic policy is optimal for the finite and infinite horizon reward criteria, respectively. It is shown that these results reduce to those derived in prior studies under the corresponding special cases, and thus may be viewed as a set of unifying optimality conditions. Numerical examples are also presented to highlight how and why an optimal policy may deviate from the otherwise-optimal myopic sensing given additional exploration opportunities, i.e., when $m<k$.

cs.IT↗

arXiv2013

Group Learning and Opinion Diffusion in a Broadcast Network

We analyze the following group learning problem in the context of opinion diffusion: Consider a network with $M$ users, each facing $N$ options. In a discrete time setting, at each time step, each user chooses $K$ out of the $N$ options, and receive randomly generated rewards, whose statistics depend on the options chosen as well as the user itself, and are unknown to the users. Each user aims to maximize their expected total rewards over a certain time horizon through an online learning process, i.e., a sequence of exploration (sampling the return of each option) and exploitation (selecting empirically good options) steps. Within this context we consider two group learning scenarios, (1) users with uniform preferences and (2) users with diverse preferences, and examine how a user should construct its learning process to best extract information from other's decisions and experiences so as to maximize its own reward. Performance is measured in {\em weak regret}, the difference between the user's total reward and the reward from a user-specific best single-action policy (i.e., always selecting the set of options generating the highest mean rewards for this user). Within each scenario we also consider two cases: (i) when users exchange full information, meaning they share the actual rewards they obtained from their choices, and (ii) when users exchange limited information, e.g., only their choices but not rewards obtained from these choices.

cs.LG↗

arXiv2013

To Stay Or To Switch: Multiuser Dynamic Channel Access

In this paper we study opportunistic spectrum access (OSA) policies in a multiuser multichannel random access cognitive radio network, where users perform channel probing and switching in order to obtain better channel condition or higher instantaneous transmission quality. However, unlikely many prior works in this area, including those on channel probing and switching policies for a single user to exploit spectral diversity, and on probing and access policies for multiple users over a single channel to exploit temporal and multiuser diversity, in this study we consider the collective switching of multiple users over multiple channels. In addition, we consider finite arrivals, i.e., users are not assumed to always have data to send and demand for channel follow a certain arrival process. Under such a scenario, the users' ability to opportunistically exploit temporal diversity (the temporal variation in channel quality over a single channel) and spectral diversity (quality variation across multiple channels at a given time) is greatly affected by the level of congestion in the system. We investigate the optimal decision process in this case, and evaluate the extent to which congestion affects potential gains from opportunistic dynamic channel switching.

eess.SY↗

arXiv2013

DorFin: WiFi Fingerprint-based Localization Revisited

Although WiFi fingerprint-based indoor localization is attractive, its accuracy remains a primary challenge especially in mobile environments. Existing approaches either appeal to physical layer information or rely on extra wireless signals for high accuracy. In this paper, we revisit the RSS fingerprint-based localization scheme and reveal crucial observations that act as the root causes of localization errors, yet are surprisingly overlooked or even unseen in previous works. Specifically, we recognize APs' diverse discrimination for fingerprinting a specific location, observe the RSS inconsistency caused by signal fluctuations and human body blockages, and uncover the RSS outdated problem on commodity smartphones. Inspired by these insights, we devise a discrimination factor to quantify different APs' discrimination, incorporate robust regression to tolerate outlier measurements, and reassemble different fingerprints to cope with outdated RSSs. Combining these techniques in a unified solution, we propose DorFin, a novel scheme of fingerprint generation, representation, and matching, which yields remarkable accuracy without incurring extra cost. Extensive experiments demonstrate that DorFin achieves mean error of 2 meters and more importantly, bounds the 95th percentile error under 5.5 meters; these are about 56% and 69% lower, respectively, compared with the state-of-the-art schemes such as Horus and RADAR.

cs.NI↗

arXiv2013

Incentives, Quality, and Risks: A Look Into the NSF Proposal Review Pilot

The National Science Foundation (NSF) will be experimenting with a new distributed approach to reviewing proposals, whereby a group of principal investigators (PIs) or proposers in a subfield act as reviewers for the proposals submitted by the same set of PIs. To encourage honesty, PIs' chances for getting funded are tied to the quality of their reviews (with respect to the reviews provided by the entire group), in addition to the quality of their proposals. Intuitively, this approach can more fairly distribute the review workload, discourage frivolous proposal submission, and encourage high quality reviews. On the other hand, this method has already raised concerns about the integrity of the process and the possibility of strategic manipulation. In this paper, we take a closer look at three specific issues in an attempt to gain a better understanding of the strengths and limitations of the new process beyond first impressions and anecdotal evidence. We start by considering the benefits and drawbacks of bundling the quality of PIs' reviews with the scientific merit of their proposals. We then consider the issue of collusion and favoritism. Finally, we examine whether the new process puts controversial proposals at a disadvantage. We conclude that some benefits of using review quality as an incentive mechanism may outweigh its drawbacks. On the other hand, even a coalition of two PIs can cause significant harm to the process, as the built-in incentives are not strong enough to deter collusion. While we also confirm the common suspicion that the process is skewed toward non-controversial proposals, the more unexpected finding is that among equally controversial proposals, those of lower quality get a leg up through this process. Thus the process not only favors non-controversial proposals, but in some sense, mediocrity. We also discuss possible ways to improve this review process.

cs.GT↗

arXiv2013

Perceptions and Truth: A Mechanism Design Approach to Crowd-Sourcing Reputation

We consider a distributed multi-user system where individual entities possess observations or perceptions of one another, while the truth is only known to themselves, and they might have an interest in withholding or distorting the truth. We ask the question whether it is possible for the system as a whole to arrive at the correct perceptions or assessment of all users, referred to as their reputation, by encouraging or incentivizing the users to participate in a collective effort without violating private information and self-interest. Two specific applications, online shopping and network reputation, are provided to motivate our study and interpret the results. In this paper we investigate this problem using a mechanism design theoretic approach. We introduce a number of utility models representing users' strategic behavior, each consisting of one or both of a truth element and an image element, reflecting the user's desire to obtain an accurate view of the other and an inflated image of itself. For each model, we either design a mechanism that achieves the optimal performance (solution to the corresponding centralized problem), or present individually rational sub-optimal solutions. In the latter case, we demonstrate that even when the centralized solution is not achievable, by using a simple punish-reward mechanism, not only a user has the incentive to participate and provide information, but also that this information can improve the system performance.

cs.GT↗

arXiv2013

Online Learning in a Contract Selection Problem

In an online contract selection problem there is a seller which offers a set of contracts to sequentially arriving buyers whose types are drawn from an unknown distribution. If there exists a profitable contract for the buyer in the offered set, i.e., a contract with payoff higher than the payoff of not accepting any contracts, the buyer chooses the contract that maximizes its payoff. In this paper we consider the online contract selection problem to maximize the sellers profit. Assuming that a structural property called ordered preferences holds for the buyer's payoff function, we propose online learning algorithms that have sub-linear regret with respect to the best set of contracts given the distribution over the buyer's type. This problem has many applications including spectrum contracts, wireless service provider data plans and recommendation systems.

cs.LG↗

arXiv2012

Online Learning in Decentralized Multiuser Resource Sharing Problems

In this paper, we consider the general scenario of resource sharing in a decentralized system when the resource rewards/qualities are time-varying and unknown to the users, and using the same resource by multiple users leads to reduced quality due to resource sharing. Firstly, we consider a user-independent reward model with no communication between the users, where a user gets feedback about the congestion level in the resource it uses. Secondly, we consider user-specific rewards and allow costly communication between the users. The users have a cooperative goal of achieving the highest system utility. There are multiple obstacles in achieving this goal such as the decentralized nature of the system, unknown resource qualities, communication, computation and switching costs. We propose distributed learning algorithms with logarithmic regret with respect to the optimal allocation. Our logarithmic regret result holds under both i.i.d. and Markovian reward models, as well as under communication, computation and switching costs.

cs.LG↗

arXiv2012

Dynamic Pricing of Power in Smart-Grid Networks

In this paper we introduce the problem of dynamic pricing of power for smart-grid networks. This is studied within a network utility maximization (NUM) framework in a deterministic setting with a single provider, multiple users and a finite horizon. The provider produces power or buys power in a (deterministic) spot market, and determines a dynamic price to charge the users. The users then adjust their demand in response to the time-varying prices. This is typically categorized as the demand response problem, and we study a progression of related models by focusing on two aspects: 1) the characterization of the structure of the optimal dynamic prices in the Smart Grid and the optimal demand and supply under various interaction with a spot market; 2) a greedy approach to facilitate the solution process of the aggregate NUM problem and the optimality gap between the greedy solution and the optimal one.

eess.SY↗

arXiv2012

Profit Incentive In A Secondary Spectrum Market: A Contract Design Approach

In this paper we formulate a contract design problem where a primary license holder wishes to profit from its excess spectrum capacity by selling it to potential secondary users/buyers. It needs to determine how to optimally price the excess spectrum so as to maximize its profit, knowing that this excess capacity is stochastic in nature, does not come with exclusive access, and cannot provide deterministic service guarantees to a buyer. At the same time, buyers are of different {\em types}, characterized by different communication needs, tolerance for the channel uncertainty, and so on, all of which a buyer's private information. The license holder must then try to design different contracts catered to different types of buyers in order to maximize its profit. We address this problem by adopting as a reference a traditional spectrum market where the buyer can purchase exclusive access with fixed/deterministic guarantees. We fully characterize the optimal solution in the cases where there is a single buyer type, and when multiple types of buyers share the same, known channel condition as a result of the primary user activity. In the most general case we construct an algorithm that generates a set of contracts in a computationally efficient manner, and show that this set is optimal when the buyer types satisfy a monotonicity condition.

cs.CE↗

arXiv2012

Throughput Optimal Switching in Multi-channel WLANs

We observe that in a multi-channel wireless system, an opportunistic channel/spectrum access scheme that solely focuses on channel quality sensing measured by received SNR may induce users to use channels that, while providing better signals, are more congested. Ultimately the notion of channel quality should include both the signal quality and the level of congestion, and a good multi-channel access scheme should take both into account in deciding which channel to use and when. Motivated by this, we focus on the congestion aspect and examine what type of dynamic channel switching schemes may result in the best system throughput performance. Specifically we derive the stability region of a multi-user multi-channel WLAN system and determine the throughput optimal channel switching scheme within a certain class of schemes.

eess.SY↗

arXiv2011

Online Learning for Combinatorial Network Optimization with Restless Markovian Rewards

Combinatorial network optimization algorithms that compute optimal structures taking into account edge weights form the foundation for many network protocols. Examples include shortest path routing, minimal spanning tree computation, maximum weighted matching on bipartite graphs, etc. We present CLRMR, the first online learning algorithm that efficiently solves the stochastic version of these problems where the underlying edge weights vary as independent Markov chains with unknown dynamics. The performance of an online learning algorithm is characterized in terms of regret, defined as the cumulative difference in rewards between a suitably-defined genie, and that obtained by the given algorithm. We prove that, compared to a genie that knows the Markov transition matrices and uses the single-best structure at all times, CLRMR yields regret that is polynomial in the number of edges and nearly-logarithmic in time.

cs.LG↗

arXiv2011

Performance and Convergence of Multi-user Online Learning

We study the problem of allocating multiple users to a set of wireless channels in a decentralized manner when the channel quali- ties are time-varying and unknown to the users, and accessing the same channel by multiple users leads to reduced quality due to interference. In such a setting the users not only need to learn the inherent channel quality and at the same time the best allocations of users to channels so as to maximize the social welfare. Assuming that the users adopt a certain online learning algorithm, we investigate under what conditions the socially optimal allocation is achievable. In particular we examine the effect of different levels of knowledge the users may have and the amount of communications and cooperation. The general conclusion is that when the cooperation of users decreases and the uncertainty about channel payoffs increases it becomes harder to achieve the socially opti- mal allocation.

cs.MA↗

arXiv2011

On the Combinatorial Multi-Armed Bandit Problem with Markovian Rewards

We consider a combinatorial generalization of the classical multi-armed bandit problem that is defined as follows. There is a given bipartite graph of $M$ users and $N \geq M$ resources. For each user-resource pair $(i,j)$, there is an associated state that evolves as an aperiodic irreducible finite-state Markov chain with unknown parameters, with transitions occurring each time the particular user $i$ is allocated resource $j$. The user $i$ receives a reward that depends on the corresponding state each time it is allocated the resource $j$. The system objective is to learn the best matching of users to resources so that the long-term sum of the rewards received by all users is maximized. This corresponds to minimizing regret, defined here as the gap between the expected total reward that can be obtained by the best-possible static matching and the expected total reward that can be achieved by a given algorithm. We present a polynomial-storage and polynomial-complexity-per-step matching-learning algorithm for this problem. We show that this algorithm can achieve a regret that is uniformly arbitrarily close to logarithmic in time and polynomial in the number of users and resources. This formulation is broadly applicable to scheduling and switching problems in networks and significantly extends prior results in the area.

math.OC↗

arXiv2011

Online Learning of Rested and Restless Bandits

In this paper we study the online learning problem involving rested and restless multiarmed bandits with multiple plays. The system consists of a single player/user and a set of K finite-state discrete-time Markov chains (arms) with unknown state spaces and statistics. At each time step the player can play M arms. The objective of the user is to decide for each step which M of the K arms to play over a sequence of trials so as to maximize its long term reward. The restless multiarmed bandit is particularly relevant to the application of opportunistic spectrum access (OSA), where a (secondary) user has access to a set of K channels, each of time-varying condition as a result of random fading and/or certain primary users' activities.

math.OC↗

arXiv2010

Spectrum Sharing as Spatial Congestion Games

In this paper, we present and analyze the properties of a new class of games - the spatial congestion game (SCG), which is a generalization of the classical congestion game (CG). In a classical congestion game, multiple users share the same set of resources and a user's payoff for using any resource is a function of the total number of users sharing it. As a potential game, this game enjoys some very appealing properties, including the existence of a pure strategy Nash equilibrium (NE) and that every improvement path is finite and leads to such a NE (also called the finite improvement property or FIP). While it's tempting to use this model to study spectrum sharing, it does not capture the spatial reuse feature of wireless communication, where resources (interpreted as channels) may be reused without increasing congestion provided that users are located far away from each other. This motivates us to study an extended form of the congestion game where a user's payoff for using a resource is a function of the number of its interfering users sharing it. This naturally results in a spatial congestion game (SCG), where users are placed over a network (or a conflict graph). We study fundamental properties of a spatial congestion game; in particular, we seek to answer under what conditions this game possesses the finite improvement property or a Nash equilibrium. We also discuss the implications of these results when applied to wireless spectrum sharing.

cs.GT↗