Recently, much attention has been devoted to discrete preference games to model the formation of ... more Recently, much attention has been devoted to discrete preference games to model the formation of opinions in social networks. More specifically, these games model the agents' strategic decision of expressing publicly an opinion, which is a result of an interplay between the agent's private belief and the social pressure. However, these games have very limited expressive power; they can model only very simple social relations and they assume that all the agents respond to social pressure in the same way. In this paper, we define and study the novel class of generalized discrete preference games. These games have additional characteristics that can model social relations to allies or competitors and complex relations among more than two agents. Moreover, they introduce different levels of strength for each relation, and they personalize the dependence of each agent to her neighborhood. We show that these novel games admit generalized ordinal potential functions and, more impor...
We consider the problem of scheduling jobs on related machines owned by selfish agents and provid... more We consider the problem of scheduling jobs on related machines owned by selfish agents and provide the first deterministic mechanisms with constant approximation that are truthful; that is, truth-telling is a dominant strategy for all agents. More precisely, we present deterministic polynomial-time (2 + epsilon)-approximation algorithms and suitable payment functions that yield truthful mechanisms for several NP-hard restrictions of this problem. Our result also yields a family of deterministic polynomial-time truthful (4+epsilon)-approximation mechanisms for any fixed number of machines. The only previously-known mechanism for this problem (proposed by Archer and Tardos [FOCS 2001]) is 3-approximated, randomized and truthful under a weaker notion of truthfulness. Up to our knowledge, our mechanisms are the first non-trivial polynomial-time deterministic truthful mechanisms for this NP-hard problem. To obtain our results we introduce a technique to transform the PTAS by Graham into a deterministic truthful mechanism
Opinion diffusion has been largely studied in the literature on settings where the opinion whose ... more Opinion diffusion has been largely studied in the literature on settings where the opinion whose spread has to be maximized, say white, competes against one opinion only, say black. For instance, for diffusion mechanisms modeled in terms of best response dynamics over majority agents (who change their opinion as to conform it to the majority of their neighbors), it is known that the spread can be maximized via certain greedy dynamics that can be computed in polynomial time. This setting is precisely the one considered in the paper. However, differently from earlier literature, it is assumed that one further opinion, say gray, is available to the agents. Moving from the observation that, with the third alternative to hand, greedy dynamics can dramatically fail to maximize the spread of opinion \emphwhite, the paper then embarks in thorough computational, algorithmic and experimental studies. The picture that emerges is totally different from what is known for the case when two opinio...
Security Games have been widely adopted to model scenarios in which one player, the Defender, has... more Security Games have been widely adopted to model scenarios in which one player, the Defender, has to decide how to deploy her resources to minimize the loss that can be caused by an attack performed by another player, the Attacker, aiming at maximizing such loss. In the present paper, we focus on scenarios in which the Defender has lexicographic-like preferences on the targets, being primarily interested in defending the integrity of a subset of the targets and, only secondarily, to reduce the amount of the other damaged targets. Our central motivation for studying this problem comes from the need to reduce the impact of malicious flows in networks, that can be either physical, like cities, or virtual, e.g., social networks. In this work, we introduce a new class of security games to model these scenarios, characterizing it and proving the NP-hardness of computing a leader-follower equilibrium, which is the most appropriate solution concept for this setting. To compute such an equil...
Online social networks are nowadays one of the most effective and widespread tools used to share ... more Online social networks are nowadays one of the most effective and widespread tools used to share information. In addition to being employed by individuals for communicating with friends and acquaintances, and by brands for marketing and customer service purposes, they constitute a primary source of daily news for a significant number of users. Unfortunately, besides legit news, social networks also allow to effectively spread inaccurate or even entirely fabricated ones. Also due to sensationalist claims, misinformation can spread from the original sources to a large number of users in a very short time, with negative consequences that, in extreme cases, can even put at risk public safety or health. In this work we discuss and propose methods to limit the spread of misinformation over online social networks. The issue is split in two separate sub-problems. We first aim to identify the most probable sources of the misinformation among the subset of users that have been reached by it. ...
Proceedings of the AAAI Conference on Artificial Intelligence, 2019
Friedkin and Johnsen (1990) modeled opinion formation in social networks as a dynamic process whi... more Friedkin and Johnsen (1990) modeled opinion formation in social networks as a dynamic process which evolves in rounds: at each round each agent updates her expressed opinion to a weighted average of her innate belief and the opinions expressed in the previous round by her social neighbors. The stubbornness level of an agent represents the tendency of the agent to express an opinion close to her innate belief. Motivated by the observation that innate beliefs, stubbornness levels and even social relations can co-evolve together with the expressed opinions, we present a new model of opinion formation where the dynamics runs in a co-evolving environment. We assume that agents’ stubbornness and social relations can vary arbitrarily, while their innate beliefs slowly change as a function of the opinions they expressed in the past. We prove that, in our model, the opinion formation dynamics converges to a consensus if reasonable conditions on the structure of the social relationships and o...
. In this paper we study the Graph Motion Planning of 1Robot problem (GMP1R) on a tree. This prob... more . In this paper we study the Graph Motion Planning of 1Robot problem (GMP1R) on a tree. This problem consists in computinga minimum cost plan for moving a robot from one vertex to another ina tree whose vertices can have movable obstacles.Papadimitriou et alt. [FOCS 94] introduced the problem and gave analgorithm for the GMP1R on a tree. Their approach
— Over the last years, computation and networking have been increasingly embedded into the enviro... more — Over the last years, computation and networking have been increasingly embedded into the environment. This tendency has been often referred to as pervasive or ubiquitous computing, to remark the aim to a dense and widespread interac-tion among computing devices. User intervention and awareness are discarded, in opposition to an automatic adaptation of applications to location and context. To this aim, much attention is drawn to technologies supporting dynamicity and mobility over small devices which can follow the user anytime, anywhere. The Bluetooth standard particularly fits this idea, by providing a versatile and flexible wireless network technology with low power consumption. Operating in a license-free frequency, users are neither charged for accessing the network nor they need an account with any company. Bluetooth dynamically sets up and manages evolving networks, by providing the possibility of automatically discovering devices and services within its transmission range. ...
We study combinatorial optimization problems involving one-parameter selfish agents considered by... more We study combinatorial optimization problems involving one-parameter selfish agents considered by Archer and Tardos [FOCS 2001]. In particular, we show that, if agents can lie in one direction (that is they either overbid or underbid) then any (polynomial-time) c-approximation algorithm, for the optimization problem without selfish agents, can be turned into a (polynomial-time) c(1 + ɛ)-approximation truthful mechanism, for any ɛ>0. We then look at the Q||Cmax problem in the case of agents owning machines of different speeds. We consider the model in which payments are given to the agents only after the machines have completed the jobs assigned. This means that for each machine that receives at least one job, the mechanism can verify if the corresponding agent declared a greater speed. For this setting, we characterize the allocation algorithms A that admit a payment function P such that M =(A, P) is a truthful mechanism. In addition, we give a (1 + ɛ)approximation truthful mecha...
We study logit dynamics [Blu93] for strategic games. At every stage of the game a player is selec... more We study logit dynamics [Blu93] for strategic games. At every stage of the game a player is selected uniformly at random and she plays according to a noisy best-response dynamics where the noise level is tuned by a parameter β. Such a dynamics defines a family of ergodic Markov chains, indexed by β, over the set of strategy profiles. Our aim is twofold: On the one hand, we are interested in the expected social welfare when the strategy profiles are random according to the stationary distribution of the Markov chain, because we believe it gives a meaningful description of the long-term behavior of the system. On the other hand, we want to estimate how long it takes, for a system starting at an arbitrary profile and running the logit dynamics, to get close to the stationary distribution; i.e., the mixing time of the chain. In this paper we study the stationary expected social welfare for the 3-player CK game [CK05], for 2-player coordination games (the same class of games studied in [...
Abstract. We study the online version of the scheduling problem Q||Cmax involving selfish agents,... more Abstract. We study the online version of the scheduling problem Q||Cmax involving selfish agents, considered by Archer and Tardos in [FOCS 2001], where jobs must be scheduled on m related machines, each of them owned by a different selfish agent. We present a general technique for transforming competitive online algorithms for Q||Cmax into truthful online mechanisms with a small loss of competitiveness. We also investigate the issue of designing new online algorithms from scratch so to obtain efficient competitive mechanisms, and prove some lower bounds on a class of “natural ” algorithms. A “direct ” use of such natural algorithms to construct truthful mechanisms yields only trivial upper bounds for the case of two machines. Finally, we consider mechanisms with verification introduced by Nisan and Ronen [STOC 1999] for offline scheduling problems. We present the first constant-competitive online truthful mechanism with verification for any number of machines. 1
Recently, much attention has been devoted to discrete preference games to model the formation of ... more Recently, much attention has been devoted to discrete preference games to model the formation of opinions in social networks. More specifically, these games model the agents' strategic decision of expressing publicly an opinion, which is a result of an interplay between the agent's private belief and the social pressure. However, these games have very limited expressive power; they can model only very simple social relations and they assume that all the agents respond to social pressure in the same way. In this paper, we define and study the novel class of generalized discrete preference games. These games have additional characteristics that can model social relations to allies or competitors and complex relations among more than two agents. Moreover, they introduce different levels of strength for each relation, and they personalize the dependence of each agent to her neighborhood. We show that these novel games admit generalized ordinal potential functions and, more impor...
We consider the problem of scheduling jobs on related machines owned by selfish agents and provid... more We consider the problem of scheduling jobs on related machines owned by selfish agents and provide the first deterministic mechanisms with constant approximation that are truthful; that is, truth-telling is a dominant strategy for all agents. More precisely, we present deterministic polynomial-time (2 + epsilon)-approximation algorithms and suitable payment functions that yield truthful mechanisms for several NP-hard restrictions of this problem. Our result also yields a family of deterministic polynomial-time truthful (4+epsilon)-approximation mechanisms for any fixed number of machines. The only previously-known mechanism for this problem (proposed by Archer and Tardos [FOCS 2001]) is 3-approximated, randomized and truthful under a weaker notion of truthfulness. Up to our knowledge, our mechanisms are the first non-trivial polynomial-time deterministic truthful mechanisms for this NP-hard problem. To obtain our results we introduce a technique to transform the PTAS by Graham into a deterministic truthful mechanism
Opinion diffusion has been largely studied in the literature on settings where the opinion whose ... more Opinion diffusion has been largely studied in the literature on settings where the opinion whose spread has to be maximized, say white, competes against one opinion only, say black. For instance, for diffusion mechanisms modeled in terms of best response dynamics over majority agents (who change their opinion as to conform it to the majority of their neighbors), it is known that the spread can be maximized via certain greedy dynamics that can be computed in polynomial time. This setting is precisely the one considered in the paper. However, differently from earlier literature, it is assumed that one further opinion, say gray, is available to the agents. Moving from the observation that, with the third alternative to hand, greedy dynamics can dramatically fail to maximize the spread of opinion \emphwhite, the paper then embarks in thorough computational, algorithmic and experimental studies. The picture that emerges is totally different from what is known for the case when two opinio...
Security Games have been widely adopted to model scenarios in which one player, the Defender, has... more Security Games have been widely adopted to model scenarios in which one player, the Defender, has to decide how to deploy her resources to minimize the loss that can be caused by an attack performed by another player, the Attacker, aiming at maximizing such loss. In the present paper, we focus on scenarios in which the Defender has lexicographic-like preferences on the targets, being primarily interested in defending the integrity of a subset of the targets and, only secondarily, to reduce the amount of the other damaged targets. Our central motivation for studying this problem comes from the need to reduce the impact of malicious flows in networks, that can be either physical, like cities, or virtual, e.g., social networks. In this work, we introduce a new class of security games to model these scenarios, characterizing it and proving the NP-hardness of computing a leader-follower equilibrium, which is the most appropriate solution concept for this setting. To compute such an equil...
Online social networks are nowadays one of the most effective and widespread tools used to share ... more Online social networks are nowadays one of the most effective and widespread tools used to share information. In addition to being employed by individuals for communicating with friends and acquaintances, and by brands for marketing and customer service purposes, they constitute a primary source of daily news for a significant number of users. Unfortunately, besides legit news, social networks also allow to effectively spread inaccurate or even entirely fabricated ones. Also due to sensationalist claims, misinformation can spread from the original sources to a large number of users in a very short time, with negative consequences that, in extreme cases, can even put at risk public safety or health. In this work we discuss and propose methods to limit the spread of misinformation over online social networks. The issue is split in two separate sub-problems. We first aim to identify the most probable sources of the misinformation among the subset of users that have been reached by it. ...
Proceedings of the AAAI Conference on Artificial Intelligence, 2019
Friedkin and Johnsen (1990) modeled opinion formation in social networks as a dynamic process whi... more Friedkin and Johnsen (1990) modeled opinion formation in social networks as a dynamic process which evolves in rounds: at each round each agent updates her expressed opinion to a weighted average of her innate belief and the opinions expressed in the previous round by her social neighbors. The stubbornness level of an agent represents the tendency of the agent to express an opinion close to her innate belief. Motivated by the observation that innate beliefs, stubbornness levels and even social relations can co-evolve together with the expressed opinions, we present a new model of opinion formation where the dynamics runs in a co-evolving environment. We assume that agents’ stubbornness and social relations can vary arbitrarily, while their innate beliefs slowly change as a function of the opinions they expressed in the past. We prove that, in our model, the opinion formation dynamics converges to a consensus if reasonable conditions on the structure of the social relationships and o...
. In this paper we study the Graph Motion Planning of 1Robot problem (GMP1R) on a tree. This prob... more . In this paper we study the Graph Motion Planning of 1Robot problem (GMP1R) on a tree. This problem consists in computinga minimum cost plan for moving a robot from one vertex to another ina tree whose vertices can have movable obstacles.Papadimitriou et alt. [FOCS 94] introduced the problem and gave analgorithm for the GMP1R on a tree. Their approach
— Over the last years, computation and networking have been increasingly embedded into the enviro... more — Over the last years, computation and networking have been increasingly embedded into the environment. This tendency has been often referred to as pervasive or ubiquitous computing, to remark the aim to a dense and widespread interac-tion among computing devices. User intervention and awareness are discarded, in opposition to an automatic adaptation of applications to location and context. To this aim, much attention is drawn to technologies supporting dynamicity and mobility over small devices which can follow the user anytime, anywhere. The Bluetooth standard particularly fits this idea, by providing a versatile and flexible wireless network technology with low power consumption. Operating in a license-free frequency, users are neither charged for accessing the network nor they need an account with any company. Bluetooth dynamically sets up and manages evolving networks, by providing the possibility of automatically discovering devices and services within its transmission range. ...
We study combinatorial optimization problems involving one-parameter selfish agents considered by... more We study combinatorial optimization problems involving one-parameter selfish agents considered by Archer and Tardos [FOCS 2001]. In particular, we show that, if agents can lie in one direction (that is they either overbid or underbid) then any (polynomial-time) c-approximation algorithm, for the optimization problem without selfish agents, can be turned into a (polynomial-time) c(1 + ɛ)-approximation truthful mechanism, for any ɛ>0. We then look at the Q||Cmax problem in the case of agents owning machines of different speeds. We consider the model in which payments are given to the agents only after the machines have completed the jobs assigned. This means that for each machine that receives at least one job, the mechanism can verify if the corresponding agent declared a greater speed. For this setting, we characterize the allocation algorithms A that admit a payment function P such that M =(A, P) is a truthful mechanism. In addition, we give a (1 + ɛ)approximation truthful mecha...
We study logit dynamics [Blu93] for strategic games. At every stage of the game a player is selec... more We study logit dynamics [Blu93] for strategic games. At every stage of the game a player is selected uniformly at random and she plays according to a noisy best-response dynamics where the noise level is tuned by a parameter β. Such a dynamics defines a family of ergodic Markov chains, indexed by β, over the set of strategy profiles. Our aim is twofold: On the one hand, we are interested in the expected social welfare when the strategy profiles are random according to the stationary distribution of the Markov chain, because we believe it gives a meaningful description of the long-term behavior of the system. On the other hand, we want to estimate how long it takes, for a system starting at an arbitrary profile and running the logit dynamics, to get close to the stationary distribution; i.e., the mixing time of the chain. In this paper we study the stationary expected social welfare for the 3-player CK game [CK05], for 2-player coordination games (the same class of games studied in [...
Abstract. We study the online version of the scheduling problem Q||Cmax involving selfish agents,... more Abstract. We study the online version of the scheduling problem Q||Cmax involving selfish agents, considered by Archer and Tardos in [FOCS 2001], where jobs must be scheduled on m related machines, each of them owned by a different selfish agent. We present a general technique for transforming competitive online algorithms for Q||Cmax into truthful online mechanisms with a small loss of competitiveness. We also investigate the issue of designing new online algorithms from scratch so to obtain efficient competitive mechanisms, and prove some lower bounds on a class of “natural ” algorithms. A “direct ” use of such natural algorithms to construct truthful mechanisms yields only trivial upper bounds for the case of two machines. Finally, we consider mechanisms with verification introduced by Nisan and Ronen [STOC 1999] for offline scheduling problems. We present the first constant-competitive online truthful mechanism with verification for any number of machines. 1
Uploads
Papers by Vincenzo Auletta