arXiv subjects

Ioannis Avramopoulos

Publications and source records attributed to Ioannis Avramopoulos.

13 recordsLinked to original sources

arXiv2025

Multiplicative weights, equalizers, and P=PPAD

We show that, by using multiplicative weights in a game-theoretic thought experiment (and an important convexity result on the composition of multiplicative weights with the relative entropy function), a symmetric bimatrix game (that is, a bimatrix matrix wherein the payoff matrix of each player is the transpose of the payoff matrix of the other) either has an interior symmetric equilibrium or there is a pure strategy that is weakly dominated by some mixed strategy. Weakly dominated pure strategies can be detected and eliminated in polynomial time by solving a linear program. Furthermore, interior symmetric equilibria are a special case of a more general notion, namely, that of an "equalizer," which can also be computed efficiently in polynomial time by solving a linear program. An elegant "symmetrization method" of bimatrix games [Jurg et al., 1992] and the well-known PPAD-completeness results on equilibrium computation in bimatrix games [Daskalakis et al., 2009, Chen et al., 2009] imply then the compelling P = PPAD.

cs.GT↗

arXiv2024

On the tractability of Nash equilibrium

In this paper, we propose a method for solving a PPAD-complete problem [Papadimitriou, 1994]. Given is the payoff matrix $C$ of a symmetric bimatrix game $(C, C^T)$ and our goal is to compute a Nash equilibrium of $(C, C^T)$. In this paper, we devise a nonlinear replicator dynamic (whose right-hand-side can be obtained by solving a pair of convex optimization problems) with the following property: Under any invertible $0 < C \leq 1$, every orbit of our dynamic starting at an interior strategy of the standard simplex approaches a set of strategies of $(C, C^T)$ such that, for each strategy in this set, a symmetric Nash equilibrium strategy can be computed by solving the aforementioned convex mathematical programs. We prove convergence using results in analysis (the analytic implicit function theorem), nonlinear optimization theory (duality theory, Berge's maximum principle, and a theorem of Robinson [1980] on the Lipschitz continuity of parametric nonlinear programs), and dynamical systems theory (such as the LaSalle invariance principle).

math.OC↗

arXiv2023

On algorithmically boosting fixed-point computations

The main topic of this paper are algorithms for computing Nash equilibria. We cast our particular methods as instances of a general algorithmic abstraction, namely, a method we call {\em algorithmic boosting}, which is also relevant to other fixed-point computation problems. Algorithmic boosting is the principle of computing fixed points by taking (long-run) averages of iterated maps and it is a generalization of exponentiation. We first define our method in the setting of nonlinear maps. Secondly, we restrict attention to convergent linear maps (for computing dominant eigenvectors, for example, in the PageRank algorithm) and show that our algorithmic boosting method can set in motion {\em exponential speedups in the convergence rate}. Thirdly, we show that algorithmic boosting can convert a (weak) non-convergent iterator to a (strong) convergent one. We also consider a {\em variational approach} to algorithmic boosting providing tools to convert a non-convergent continuous flow to a convergent one. Then, by embedding the construction of averages in the design of the iterated map, we constructively prove the existence of Nash equilibria (and, therefore, Brouwer fixed points). We then discuss implementations of averaging and exponentiation, an important matter even for the scalar case. We finally discuss a relationship between dominant (PageRank) eigenvectors and Nash equilibria.

cs.GT↗

arXiv2023

A dynamic that evolves toward a Nash equilibrium

In this paper, we study an exponentiated multiplicative weights dynamic based on Hedge, a well-known algorithm in theoretical machine learning and algorithmic game theory. The empirical average (arithmetic mean) of the iterates Hedge generates is known to approach a minimax equilibrium in zero-sum games. We generalize that result to show that a weighted version of the empirical average converges to an equilibrium in the class of symmetric bimatrix games for a diminishing learning rate parameter. Our dynamic is the first dynamical system (whether continuous or discrete) shown to evolve toward a Nash equilibrium without assuming monotonicity of the payoff structure or that a potential function exists. Although our setting is somewhat restricted, it is also general as the class of symmetric bimatrix games captures the entire computational complexity of the PPAD class (even to approximate an equilibrium).

cs.GT↗

arXiv2021

Poising on Ariadne's thread: An algorithm for computing a maximum clique in polynomial time

In this paper, we present a polynomial-time algorithm for the maximum clique problem, which implies P = NP. Our algorithm is based on a continuous game-theoretic representation of this problem and at its heart lies a discrete-time dynamical system. The rule of our dynamical system depends on a parameter such that if this parameter is equal to the maximum-clique size, the iterates of our dynamical system are guaranteed to converge to a maximum clique.

cs.GT↗

arXiv2020

Evolution toward a Nash equilibrium

In this paper, we study the dynamic behavior of Hedge, a well-known algorithm in theoretical machine learning and algorithmic game theory. The empirical average (arithmetic mean) of the iterates Hedge generates is known to converge to a minimax equilibrium in zero-sum games. We generalize that result to show convergence of the empirical average to Nash equilibrium in symmetric bimatrix games (that is bimatrix games where the payoff matrix of each player is the transpose of that of the other) in the sense that every limit point of the sequence of averages is an $ε$-approximate symmetric equilibrium strategy for any desirable $ε$. Our analysis gives rise to a symmetric equilibrium fully polynomial-time approximation scheme, implying P = PPAD.

cs.GT↗

arXiv2018

On incremental deployability

Motivated by the difficulty of effecting fundamental change in the architecture of the Internet, in this paper, we study from a theoretical perspective the question of how individuals can join forces toward collective ventures. To that end, we draw on an elementary concept in Internet systems engineering, namely, that of incremental deployability, which we study mathematically and computationally. For example, we show that incremental deployability is at least as general a concept as the Nash equilibrium (in that the latter can be derived from the former). We then draw on this foundation to design and analyze institutional mechanisms that are not only promising to bootstrap emerging Internet architectures but they also have broader applications in social organization beyond its predominant market (and finance)-based character.

cs.GT↗

arXiv2016

Evolutionary stability implies asymptotic stability under multiplicative weights

We show that evolutionarily stable states in general (nonlinear) population games (which can be viewed as continuous vector fields constrained on a polytope) are asymptotically stable under a multiplicative weights dynamic (under appropriate choices of a parameter called the learning rate or step size, which we demonstrate to be crucial to achieve convergence, as otherwise even chaotic behavior is possible to manifest). Our result implies that evolutionary theories based on multiplicative weights are compatible (in principle, more general) with those based on the notion of evolutionary stability. However, our result further establishes multiplicative weights as a nonlinear programming primitive (on par with standard nonlinear programming methods) since various nonlinear optimization problems, such as finding Nash/Wardrop equilibria in nonatomic congestion games, which are well-known to be equipped with a convex potential function, and finding strict local maxima of quadratic programming problems, are special cases of the problem of computing evolutionarily stable states in nonlinear population games.

cs.GT↗

arXiv2016

Jump-starting coordination in a stag hunt: Motivation, mechanisms, and their analysis

The stag hunt (or assurance game) is a simple game that has been used as a prototype of a variety of social coordination problems (ranging from the social contract to the adoption of technical standards). Players have the option to either use a superior cooperative strategy whose payoff depends on the other players' choices or use an inferior strategy whose payoff is independent of what other players do; the cooperative strategy may incur a loss if sufficiently many other players do not cooperate. Stag hunts have two (strict) pure Nash equilibria, namely, universal cooperation and universal defection (as well as a mixed equilibrium of low predictive value). Selection of the inferior (pure) equilibrium is called a coordination failure. In this paper, we present and analyze using game-theoretic techniques mechanisms aiming to avert coordination failures and incite instead selection of the superior equilibrium. Our analysis is based on the solution concepts of Nash equilibrium, dominance solvability, as well as a formalization of the notion of "incremental deployability," which is shown to be keenly relevant to the sink equilibrium.

cs.GT↗

arXiv2015

On the computational complexity of evolution

It is well-known that the problem of recognizing an ESS in a symmetric bimatrix game is coNP-complete. In this paper, we show that recognizing an ESS even in doubly symmetric bimatrix games is also coNP-complete. Our result further implies that recognizing asymptotically stable equilibria of the replicator dynamic in this class of games is also a coNP-complete problem.

cs.CC↗

arXiv2014

Multiplicative weights in monotropic games

We introduce a new class of population games that we call monotropic; these are games characterized by the presence of a unique globally neutrally stable Nash equilibrium. Monotropic games generalize strictly concave potential games and zero sum games with a unique minimax solution. Within the class of monotropic games, we study a multiplicative weights dynamic. We show that, depending on a parameter called the learning rate, multiplicative weights are interior globally convergent to the unique equilibrium of monotropic games, but may also induce chaotic behavior if the learning rate is not carefully chosen.

cs.GT↗

arXiv2013

A general theory of equilibrium behavior

Economists were content with the concept of the Nash equilibrium as game theory's solution concept until Daskalakis, Goldberg, and Papadimitriou showed that finding a Nash equilibrium is most likely a computationally hard problem, a result that set off a deep scientific crisis. Motivated, in part, by their result, in this paper, we propose a general theory of equilibrium behavior in vector fields (and, therefore, also noncooperative games). Our line of discourse is to show that these universal in nature mathematical objects are endowed with significant structure, which we probe to unearth atypical, previously unidentified, equilibrium behavior.

cs.GT↗

arXiv2012

Architectural innovation: A game-theoretic approach

In this paper, we view the Internet under a game-theoretic lens in an effort to explain and overcome the Internet's innovation slump. Game Theory is used to model Internet environments as problems of technological competition toward the end of understanding their emergent phenomena and the evolutionary forces that shape them. However, our results extend beyond understanding the Internet architecture toward helping the Internet population achieve socially desirable outcomes.

cs.NI↗