Abstract
A model of population genetics of the Lotka–Volterra type with mutations on a statistical manifold is introduced. Mutations in the model are described by diffusion on a statistical manifold with a generator in the form of a Laplace–Beltrami operator with a Fisher–Rao metric, that is, the model combines population genetics and information geometry. This model describes a generalization of the model of machine learning theory, the model of generative adversarial network (GAN), to the case of populations of generative adversarial networks. The introduced model describes the control of overfitting for generating adversarial networks.
Similar content being viewed by others
References
I. Goodfellow, J. Pouget-Abadie, M. Mirza, B. Xu, D. Warde-Farley, S. Ozair, A. Courville, and Y. Bengio, “Generative adversarial nets,” in: 28th Annual Conference on Neural Information Processing Systems 2014 (Montreal, Canada, 8–13 December, 2014, Advances in Neural Information Processing Systems, Vol. 27, Z. Ghahramani, M. Welling, C. Cortes, N. Lawrence, and K. Q. Weinberger, eds.), NIPS Foundation, La Jolla, CA (2014), pp. 2672–2680; arXiv: 1406.2661.
I. Goodfellow, “NIPS 2016 tutorial: Generative adversarial networks,” arXiv: 1701.00160.
S. V. Kozyrev, “Learning theory and population genetics,” Lobachevskii J. Math., 43, 1655–1662 (2022).
S. V. Kozyrev, “Learning by population genetics and matrix Riccati equation,” Entropy, 25, 348, 9 pp. (2023).
A. M. Turing, “Computing machinery and intelligence,” Mind, LIX, 433–460 (1950).
V. Vanchurin, Yu. I. Wolf, M. I. Katsnelson, and E. V. Koonin, “Towards a theory of evolution as multilevel learning,” Proc. Natl. Acad. Sci. USA, 119, e2120037119, 12 pp. (2022).
S. Nikolenko, A. Kadurin, and E. Arkhangelskaya, Deep Learning. Immersion in the World of Neural Networks [in Russian], Piter, St. Petersburg (2018).
M. Eigen, J. McCaskill, and P. Schuster, “Molecular quasi-species,” J. Phys. Chem., 92, 6881–6891 (1988).
E. A. Morozova and N. N. Chentsov, “Natural geometry of families of probability laws [in Russian],” in: Probability theory – 8, Itogi Nauki i Tekhniki. Ser. Sovrem. Probl. Mat. Fund. Napr., Vol. 83), VINITI, Moscow (1991), pp. 133–265.
N. N. Čencov, Statistical Decision Rules and Optimal Inferences (Translations of Mathematical Monographs, Vol. 53), AMS, Providence, RI (1972).
S. Amari, Differential-Geometrical Methods in Statistics (Lecture Notes in Statistics, Vol. 28), Springer, Berlin (1985).
S. Amari and H. Nagaoka, Methods of Information Geometry (Translations of Mathematical Monographs, Vol. 191), AMS, Providence, RI (2000).
P. Gibilisco, E. Riccomagno, M. P. Rogantin, and H. P. Wynn (eds.), Algebraic and Geometric Methods in Statistics, Cambridge Univ. Press, Cambridge (2010).
N. Combe, Yu. I. Manin, and M. Marcolli, “Geometry of information: Classical and quantum aspects,” Theor. Comput. Sci., 908, 2–27 (2022); arXiv: 2107.08006.
V. N. Vapnik, The Nature of Statistical Learning Theory, Springer, New York (2000).
S. Hochreiter and J. Schmidhuber, “Flat minima,” Neural Computation, 9, 1–42 (1997).
O. Bousquet and A. Elisseeff, “Stability and generalization,” J. Mach. Learn. Res., 2, 499–526 (2002).
S. Kutin and P. Niyogi, “Almost-everywhere algorithmic stability and generalization error,” in: Proceedings of the Eighteenth Conference on Uncertainty in Artificial Intelligence (UAI2002) (Alberta, Canada, August 1–4, 2002, A. Darwiche and N. Friedman, eds.), Morgan Kaufmann Publ., San Francisco, CA (2002), pp. 275–282; arXiv: 1301.0579.
T. Poggio, R. Rifkin, S. Mukherjee, and P. Niyogi, “General conditions for predictivity in learning theory,” Nature, 428, 419–422 (2004).
R. A. Fisher, The Genetical Theory of Natural Selection, Oxford Univ. Press, Oxford (1999).
J. M. Smith, Evolution and the Theory of Games, Cambridge Univ. Press, Cambridge (1982).
Funding
This work was supported by the Russian Science Foundation under grant No. 19-11-00320, https://rscf.ru/en/project/19-11-00320/.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
The author of this work declares that he has no conflicts of interest.
Additional information
Translated from Teoreticheskaya i Matematicheskaya Fizika, 2024, Vol. 218, pp. 320–329 https://doi.org/10.4213/tmf10533.
Publisher’s note. Pleiades Publishing remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendix
Maximum likelihood method
We consider a parametric family of probability distributions \(p(x,\theta)\), \(x\in X\), with a parameter \(\theta\). Let \(\{x_i\}\) be a sample of independent trials in \(X\) (we assume the existence of a probability distribution \(q\) on \(X\) that generates such a sample). The likelihood function has the form of the product of probability densities for events from the sample
A maximum likelihood estimate for \(\theta\),
Overfitting is a well-known problem in learning theory: fitting \(p(x,\theta)\) to the training sample \(\{x_i\}\) can have a high likelihood but a low likelihood on the control sample \(\{x'_i\}\) (some other sample generated by the distribution \(q\)). The control of overfitting in learning theory usually reduces to regularization (additions to the empirical sum that change the form of the likelihood function, or another kind of the empirical risk functional).
Game theory
A game is a function of several variables called players. A game takes values equal to a set of real numbers, one for each player:
Example: two players, a zero-sum game (the sum of the players payoffs is zero).
A mixed strategy for the \(i\)th player: the probability distribution \(p_i(s_j^i)\) for this player’s strategies, the index \(j\) enumerates different strategies for the \(i\)th player. The payoff of the \(i\)th player for a set of mixed strategies
Nash equilibrium: A set of strategies in which no participant can increase the payoff by changing their strategy if the other participants do not change their strategies. It exists in the class of mixed strategies (may not exist in the class of pure strategies).
Maximin: the largest payoff that a given player can obtain without knowing the actions of other players,
Minimax: The smallest payoff that other players can force without knowing the player’s action,
The minimax is not less than the maximin, \(\overline{v_i}\ge \underline{v_i}\).
The minimax for two-player zero-sum games is the same as the Nash equilibrium.
Lotka–Volterra model
This model describes the population dynamics of two species (predator and prey) by a system of two ODEs
The Lotka–Volterra model with mutations generalizes the above model as follows. There are prey \(x_i\) and predators \(y_m\), and the equations of population dynamics have the form
For a bounded ecological niche, the first equation takes the form
Information geometry [9]–[14]
A statistical manifold is a manifold of parameters of a parametric probability distribution, or a manifold whose points are probability distributions on \(X\) (a smooth dependence of the distribution on parameters is assumed).
The (nonsymmetric) Kullback–Leibler distance between two probability distributions on \(X\) is defined as
Rights and permissions
About this article
Cite this article
Kozyrev, S.V. Lotka–Volterra model with mutations and generative adversarial networks. Theor Math Phys 218, 276–284 (2024). https://doi.org/10.1134/S0040577924020077
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S0040577924020077