Network Learning in Quadratic Games From Best-Response Dynamics
Pages 3669 - 3684
Abstract
We investigate the capacity of an adversary to learn the underlying interaction network through repeated best response actions in linear-quadratic games. The adversary strategically perturbs the decisions of a set of action-compromised players and observes the sequential decisions of a set of action-leaked players. The central question pertains to whether such an adversary can fully reconstruct or effectively estimate the underlying interaction structure among the players. To begin with, we establish a series of results that characterize the learnability of the interaction graph from the adversary’s perspective by drawing connections between this network learning problem in games and classical system identification theory. Subsequently, taking into account the inherent stability and sparsity constraints inherent in the network interaction structure, we propose a stable and sparse system identification framework for learning the interaction graph based on complete player action observations. Moreover, we present a stable and sparse subspace identification framework for learning the interaction graph when only partially observed player actions are available. Finally, we demonstrate the efficacy of the proposed learning frameworks through numerical examples.
References
[1]
Y. Chen, K. Ding, and G. Shi, “Network learning from best-response dynamics in LQ games,” in Proc. Amer. Control Conf. (ACC), May 2023, pp. 1680–1685.
[2]
J. Von Neumann and O. Morgenstern, Theory of Games and Economic Behavior. Princeton, NJ, USA: Princeton Univ. Press, 2007.
[3]
C. Bicchieri, “Rationality and game theory,” in The Oxford Handbook of Rationality. Oxford University Press, 2004, pp. 182–205. 10.1093/oxfordhb/9780195145397.003.0010.
[4]
L. Bianco, M. Caramia, S. Giordani, and V. Piccialli, “A game-theoretic approach for regulating hazmat transportation,” Transp. Sci., vol. 50, no. 2, pp. 424–438, May 2016.
[5]
D. Niyato and E. Hossain, “Competitive pricing for spectrum sharing in cognitive radio networks: Dynamic game, inefficiency of Nash equilibrium, and collusion,” IEEE J. Sel. Areas Commun., vol. 26, no. 1, pp. 192–202, Jan. 2008.
[6]
R. Sen and R. C. King, “Revisit the debate on intermediation, disintermediation and reintermediation due to E-commerce,” Electron. Markets, vol. 13, no. 2, pp. 153–162, Jan. 2003.
[7]
M. Barreno, B. Nelson, A. D. Joseph, and J. D. Tygar, “The security of machine learning,” Mach. Learn., vol. 81, no. 2, pp. 121–148, 2010.
[8]
Y. Leng, Y. Chen, X. Dong, J. Wu, and G. Shi, “Social interaction leakages from public behavioral data: A diagnostic and differential privacy framework,” SSRN J., 2021. 10.2139/ssrn.3875878.
[9]
C. Ballester, Y. Zenou, and A. Calvó-Armengol, “Delinquent networks,” J. Eur. Econ. Assoc., vol. 8, no. 1, pp. 34–61, 2010.
[10]
A. Calvó-Armengol and Y. Zenou, “Social networks and crime decisions: The role of social structure in facilitating delinquent behavior,” Int. Econ. Rev., vol. 45, no. 3, pp. 939–958, Aug. 2004.
[11]
M. D. König, X. Liu, and Y. Zenou, “R&D networks: Theory, empirics, and policy implications,” Rev. Econ. Statist., vol. 101, no. 3, pp. 476–491, 2019.
[12]
D. Fudenberg and J. Tirole, Game Theory. Cambridge, MA, USA: MIT Press, 1991.
[13]
C. H. Papadimitriou and T. Roughgarden, “Computing correlated equilibria in multi-player games,” J. ACM, vol. 55, no. 3, pp. 1–29, Jul. 2008.
[14]
C. Ballester, A. Calvó-Armengol, and Y. Zenou, “Who’s who in networks. Wanted: The key player,” Econometrica, vol. 74, no. 5, pp. 1403–1417, Sep. 2006.
[15]
M. KEARNS, “Graphical models for game theory,” in Proc. 17th Conf. Uncertainty Artif., Intell., 2001, pp. 253–260.
[16]
J. Honorio and L. E. Ortiz, “Learning the structure and parameters of large-population graphical games from behavioral data,” J. Mach. Learn. Res., vol. 16, no. 1, pp. 1157–1210, 2015.
[17]
A. Ghoshal and J. Honorio, “Learning graphical games from behavioral data: Sufficient and necessary conditions,” in Proc. Artif. Intell. Statist., 2017, pp. 1532–1540.
[18]
Y. Leng, X. Dong, J. Wu, and A. Pentland, “Learning quadratic games on networks,” in Proc. Int. Conf. Mach. Learn., 2020, pp. 5820–5830.
[19]
A. Ghoshal and J. Honorio, “From behavior to sparse graphical games: Efficient recovery of equilibria,” in Proc. 54th Annu. Allerton Conf. Commun., Control, Comput. (Allerton), Sep. 2016, pp. 1220–1227.
[20]
V. K. Garg and T. Jaakkola, “Local aggregative games,” in Proc. 31st Int. Conf. Neural Inf. Process. Syst., 2017, pp. 5347–5357.
[21]
A. Ghoshal and J. Honorio, “Learning sparse polymatrix games in polynomial time and sample complexity,” in Proc. Int. Conf. Artif. Intell. Statist., 2018, pp. 1486–1494.
[22]
A. Barik and J. Honorio, “Provable sample complexity guarantees for learning of continuous-action graphical games with nonparametric utilities,” 2020, arXiv:2004.01022.
[23]
J. Friedman, T. Hastie, and R. Tibshirani, “Sparse inverse covariance estimation with the graphical lasso,” Biostatistics, vol. 9, no. 3, pp. 432–441, Jul. 2008.
[24]
S. K. M. Wong and C. J. Butz, “Constructing the dependency structure of a multiagent probabilistic network,” IEEE Trans. Knowl. Data Eng., vol. 13, no. 3, pp. 395–415, May 2001.
[25]
M. Gomez-Rodriguez, D. Balduzzi, and B. Schölkopf, “Uncovering the temporal dynamics of diffusion networks,” in Proc. 28th Int. Conf. Int. Conf. Mach. Learn., 2011, pp. 561–568.
[26]
X. Dong, D. Thanou, M. Rabbat, and P. Frossard, “Learning graphs from data: A signal representation perspective,” IEEE Signal Process. Mag., vol. 36, no. 3, pp. 44–63, May 2019.
[27]
K. Ouyang et al., “Fine-grained urban flow inference,” IEEE Trans. Knowl. Data Eng., vol. 34, no. 6, pp. 2755–2770, Jun. 2022.
[28]
L. Zhu, D. Guo, J. Yin, G. V. Steeg, and A. Galstyan, “Scalable temporal latent space inference for link prediction in dynamic social networks,” IEEE Trans. Knowl. Data Eng., vol. 28, no. 10, pp. 2765–2777, Oct. 2016.
[29]
H. Chen, W.-S. Ku, H. Wang, L. Tang, and M.-T. Sun, “Scaling up Markov logic probabilistic inference for social graphs,” IEEE Trans. Knowl. Data Eng., vol. 29, no. 2, pp. 433–445, Feb. 2017.
[30]
J. Zhang, C. Wang, J. Wang, J. X. Yu, J. Chen, and C. Wang, “Inferring directions of undirected social ties,” IEEE Trans. Knowl. Data Eng., vol. 28, no. 12, pp. 3276–3292, Dec. 2016.
[31]
H. Huang, Q. Yan, L. Chen, Y. Gao, and C. S. Jensen, “Statistical inference of diffusion networks,” IEEE Trans. Knowl. Data Eng., vol. 33, no. 2, pp. 742–753, Feb. 2021.
[32]
L. Hu, K. C. C. Chan, X. Yuan, and S. Xiong, “A variational Bayesian framework for cluster analysis in a complex network,” IEEE Trans. Knowl. Data Eng., vol. 32, no. 11, pp. 2115–2128, Nov. 2020.
[33]
P. Van Overschee and B. De Moor, Subspace identification for linear systems: Theory—Implementation—Applications. Cham, Switzerland: Springer, 2012.
[34]
L. Ljung, “System identification,” in Signal Analysis and Prediction. London, U.K.: Springer, 1998, pp. 163–173.
[35]
A. S. Bazanella, M. Gevers, and J. M. Hendrickx, “Network identification with partial excitation and measurement,” in Proc. IEEE 58th Conf. Decis. Control (CDC), Dec. 2019, pp. 5500–5506.
[36]
C. Yu, L. Ljung, A. Wills, and M. Verhaegen, “Constrained subspace method for the identification of structured state-space models (COSMOS),” IEEE Trans. Autom. Control, vol. 65, no. 10, pp. 4201–4214, Oct. 2020.
[37]
Y. Shen, B. Baingana, and G. B. Giannakis, “Kernel-based structural equation models for topology identification of directed networks,” IEEE Trans. Signal Process., vol. 65, no. 10, pp. 2503–2516, May 2017.
[38]
S. Oymak and N. Ozay, “Revisiting Ho–Kalman-based system identification: Robustness and finite-sample analysis,” IEEE Trans. Autom. Control, vol. 67, no. 4, pp. 1914–1928, Apr. 2022.
[39]
Y. Sattar, S. Oymak, and N. Ozay, “Finite sample identification of bilinear dynamical systems,” in Proc. IEEE 61st Conf. Decis. Control (CDC), Dec. 2022, pp. 6705–6711.
[40]
A. Chiuso and G. Pillonetto, “System identification: A machine learning perspective,” Annu. Rev. Control, Robot., Auto. Syst., vol. 2, no. 1, pp. 281–304, May 2019.
[41]
S. Ross and J. A. Bagnell, “Agnostic system identification for model-based reinforcement learning,” in Proc. 29th Int. Coference Int. Conf. Mach. Learn., 2012, pp. 1905–1912.
[42]
E. de la Rosa and W. Yu, “Randomized algorithms for nonlinear system identification with deep learning modification,” Inf. Sci., vols. 364–365, pp. 197–212, Oct. 2016.
[43]
W. Wonham, Linear Multivariable Control: A Geometric Approach, vol. 10. New York, NY, USA: Springer-Verlag, 1985.
[44]
M. O. Jackson and Y. Zenou, “Games on networks,” in Handbook of Game Theory With Economic Applications, vol. 4. Amsterdam, The Netherlands: Elsevier, 2015, pp. 95–163.
[45]
M. L. Littman, “Markov games as a framework for multi-agent reinforcement learning,” in Machine Learning Proceedings 1994. Amsterdam, The Netherlands: Elsevier, 1994, pp. 157–163.
[46]
D. Fudenberg, F. Drew, D. K. Levine, and D. K. Levine, The Theory of Learning in Games, vol. 2. Cambridge, MA, USA: MIT Press, 1998.
[47]
M. O. Jackson, Social and Economic Networks. Princeton, NJ, USA: Princeton Univ. Press, 2010.
[48]
R. Dey, Z. Jelveh, and K. Ross, “Facebook users have become much more private: A large-scale study,” in Proc. IEEE Int. Conf. Pervasive Comput. Commun. Workshops, Mar. 2012, pp. 346–352.
[49]
K. Ding and J. Zhang, “Multi-party privacy conflict management in online social networks: A network game perspective,” IEEE/ACM Trans. Netw., vol. 28, no. 6, pp. 2685–2698, Dec. 2020.
[50]
A. Calvó-Armengol, E. Patacchini, and Y. Zenou, “Peer effects and social networks in education,” Rev. Econ. Stud., vol. 76, no. 4, pp. 1239–1267, Oct. 2009.
[51]
R. W. Helsley and Y. Zenou, “Social networks and interactions in cities,” J. Econ. Theory, vol. 150, pp. 426–466, Mar. 2014.
[52]
H. J. van Waarde, P. Tesi, and M. K. Camlibel, “Topology identification of heterogeneous networks: Identifiability and reconstruction,” Automatica, vol. 123, Jan. 2021, Art. no.
[53]
P. E. Paré, V. Chetty, and S. Warnick, “On the necessity of full-state measurement for state-space network reconstruction,” in Proc. IEEE Global Conf. Signal Inf. Process., Dec. 2013, pp. 803–806.
[54]
G. Ramos, A. P. Aguiar, and S. Pequito, “An overview of structural systems theory,” Automatica, vol. 140, Jun. 2022, Art. no.
[55]
J.-M. Dion, C. Commault, and J. van der Woude, “Generic properties and control of linear structured systems: A survey,” Automatica, vol. 39, no. 7, pp. 1125–1144, Jul. 2003.
[56]
B. Cassidy, C. Rae, and V. Solo, “Brain activity: Connectivity, sparsity, and mutual information,” IEEE Trans. Med. Imag., vol. 34, no. 4, pp. 846–860, Apr. 2015.
[57]
N. Wisitpongphan, F. Bai, P. Mudalige, V. Sadekar, and O. Tonguz, “Routing in sparse vehicular ad hoc wireless networks,” IEEE J. Sel. Areas Commun., vol. 25, no. 8, pp. 1538–1556, Oct. 2007.
[58]
W. Jongeneel, T. Sutter, and D. Kuhn, “Efficient learning of a linear dynamical system with stability guarantees,” IEEE Trans. Autom. Control, vol. 68, no. 5, pp. 2790–2804, May 2023.
[59]
T. Fawcett, “An introduction to ROC analysis,” Pattern Recognit. Lett., vol. 27, no. 8, pp. 861–874, Jun. 2006.
[60]
M. O. Jackson, T. Rodriguez-Barraquer, and X. Tan, “Social capital and social quilts: Network patterns of favor exchange,” Amer. Econ. Rev., vol. 102, no. 5, pp. 1857–1897, Aug. 2012.
[61]
J. F. Padgett and C. K. Ansell, “Robust action and the rise of the medici, 1400-1434,” Amer. J. Sociol., vol. 98, no. 6, pp. 1259–1319, May 1993.
[62]
C. C. MacDuffee, The Theory Matrices, vol. 5. Cham, Switzerland: Springer, 2012.
[63]
M. Grewal and K. Glover, “Identifiability of linear and nonlinear dynamical systems,” IEEE Trans. Autom. Control, vol. AC-21, no. 6, pp. 833–837, Dec. 1976.
Index Terms
- Network Learning in Quadratic Games From Best-Response Dynamics
Index terms have been assigned to the content through auto-classification.
Recommendations
Dynamics in network interaction games
DISC'09: Proceedings of the 23rd international conference on Distributed computingWe study the convergence times of dynamics in games involving graphical relationships of players. Our model of local interaction games generalizes a variety of recently studied games in game theory and distributed computing. In a local interaction game ...
On Best-Response Dynamics in Potential Games
The paper studies the convergence properties of (continuous-time) best-response dynamics from game theory. Despite their fundamental role in game theory, best-response dynamics are poorly understood in many games of interest due to the discontinuous, set-...
Comments
Information & Contributors
Information
Published In
1063-6692 © 2024 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See https://www.ieee.org/publications/rights/index.html for more information.
Publisher
IEEE Press
Publication History
Published: 06 June 2024
Published in TON Volume 32, Issue 5
Qualifiers
- Research-article
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- 0Total Citations
- 22Total Downloads
- Downloads (Last 12 months)22
- Downloads (Last 6 weeks)16
Reflects downloads up to 28 Dec 2024
Other Metrics
Citations
View Options
Login options
Check if you have access through your login credentials or your institution to get full access on this article.
Sign in