Game on Random Environment, Mean-Field Langevin System, and Neural Networks

References

• [1] Arjovsky M, Chintala S, Bottou L (2017) Wasserstein Generative Adversarial Networks. Proc. 34th Internat. Conf. Machine Learn. Sydney, NSW, Australia, vol. 70 (JMLR.org), 214–223.Google Scholar
• [2] Aumann R (1974) Subjectivity and correlation in randomized strategies. J. Math. Econom. 1(1):67–96.Crossref, Google Scholar
• [3] Cardaliaguet P, Delarue F, Lasry JM, Lions PL (2015) The master equation and the convergence problem in mean field games. Preprint, submitted September 8, https://arxiv.org/abs/1509.02505.Google Scholar
• [4] Carmona R, Delarue F (2018) Probabilistic Theory of Mean Field Games with Applications II (Springer, Berlin).Crossref, Google Scholar
• [5] Cesa-Bianchi N, Lugosi G (2006) Prediction, Learning, and Games (Cambridge University Press, Cambridge, UK).Crossref, Google Scholar
• [6] Chassagneux JF, Szpruch L, Tse A (2019) Weak quantitative propagation of chaos via differential calculus on the space of measures. Preprint, submitted January 8, https://arxiv.org/abs/1901.02556.Google Scholar
• [7] Chen RTQ, Rubanova Y, Bettencourt J, Duvenaud D (2018) Neural ordinary differential equations. Bengio S, Wallach H, Larochelle H, Grauman K, Cesa-Bianchi N, Garnett R, eds. Proc. 32nd Internat. Conf. on Neural Inform. Processing Systems, vol. 31 (Curran Associates, Inc., Red Hook, NY).Google Scholar
• [8] Chizat L, Bach F (2018) On the global convergence of gradient descent for over-parameterized models using optimal transport. Bengio S, Wallach H, Larochelle H, Grauman K, Cesa-Bianchi N, Garnett R, eds. Adv. Neural Inform. Processing Systems, vol. 31 (Curran Associates, Inc., Red Hook, NY), 3040–3050.Google Scholar
• [9] Cuchiero C, Larsson M, Teichmann J (2020) Deep neural networks, generic universal interpolation, and controlled ODEs. SIAM Journal on Mathematics of Data Science 2(3):901–919.Google Scholar
• [10] Delarue F, Lacker D, Ramanan K (2019) From the master equation to mean field game limit theory: A central limit theorem. Electronic J. Probab. 24(51).Google Scholar
• [11] Delarue F, Lacker D, Ramanan K (2022) From the master equation to mean field game limit theory: Large deviations and concentration of measure. Preprint, submitted April 23, https://arxiv.org/abs/1804.08550.Google Scholar
• [12] Durmus A, Moulines E (2017) Nonasymptotic convergence analysis for the unadjusted langevin algorithm. Ann. Appl. Probab. 27(3):1551–1587.Crossref, Google Scholar
• [13] E W, Han J, Li Q (2019) A mean-field optimal control formulation of deep learning. Res. Math. Sci. 6:Article 10.Crossref, Google Scholar
• [14] Eberle A (2016) Reflection couplings and contraction rates for diffusions. Probab. Theory Related Fields 166(3-4):851–886.Crossref, Google Scholar
• [15] Eberle A, Guillin A, Zimmer R (2019) Quantitative Harris-type theorems for diffusions and McKean–Vlasov processes. Trans. Amer. Math. Soc. 371(10):7135–7173.Crossref, Google Scholar
• [16] Fudenberg D, Levine DK (1998) The Theory of Learning in Games (MIT Press, Cambridge, MA).Google Scholar
• [17] Gelman A, Roberts G, Gilks W (1996) Efficient metropolis jumping rules. Bernardo JM, Berger JO, Dawid AP, Smith AFM, eds. Bayesian Statistics, (Oxford University Press, Oxford), 599–608.Google Scholar
• [18] Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A, et al. (2014) Generative adversarial nets. Ghahramani Z, Welling M, Cortes C, Lawrence N, Weinberger KQ, eds. Adv. Neural Inform. Processing Systems, vol. 27 (Curran Associates, Inc., Red Hook, NY).Google Scholar
• [19] Henry D (1981) Geometric Theory of Semilinear Parabolic Equations (Springer, Berlin).Crossref, Google Scholar
• [20] Hu K, Kazeykina A, Ren Z (2019) Mean-field langevin system, optimal control and deep neural networks. Preprint, submitted September 16, https://arxiv.org/abs/1909.07278.Google Scholar
• [21] Hu K, Ren Z, Siska D, Szpruch L (2021) Mean-field Langevin dynamics and energy landscape of neural networks. Ann. Inst. Henri Poincare Probab. Statist. 57(4):2043–2065.Google Scholar
• [22] Jabir JF, Siska D, Szpruch L (2020) Mean-field neural odes via relaxed optimal control. Preprint, submitted December 11, https://arxiv.org/abs/1912.05475.Google Scholar
• [23] Jordan R, Kinderlehrer D, Otto F (1998) The variational formulation of the Fokker–Planck equation. SIAM J. Math. Anal. 29(1):1–17.Crossref, Google Scholar
• [24] Lasry JM, Lions PL (2007) Mean field games. Jpn. J. Math. 2(1):229–260.Crossref, Google Scholar
• [25] Liu H, Markowich P (2020) Selection dynamics for deep neural networks. J. Differential Equations 269(12):11540–11574.Google Scholar
• [26] Lu Y, Ma C, Lu Y, Lu J, Ying L (2020) A mean-field analysis of deep resnet and beyond: Toward provable optimization via overparameterization from depth. III HD, Singh A, eds. Proc. 37th Internat. Conf. Machine Learn. (PMLR), 119:6426–6436.Google Scholar
• [27] Mei S, Misiakiewicz T, Montanari A (2019) Mean-field theory of two-layers neural networks: dimension-free bounds and kernel limit. Proc. Machine Learn. Res. 99:1–77.Google Scholar
• [28] Mei S, Montanari A, Nguyen PM (2018) A mean field view of the landscape of two-layer neural networks. Proc. National Acad. Sci. USA 115(33):E7665–E7671.Crossref, Google Scholar
• [29] Méléard S (1996) Asymptotic behaviour of some interacting particle systems; Mckean–Vlasov and Boltzmann models. Probabilistic models for nonlinear partial differential equations (Montecatini Terme, 1995). Lecture Notes Math. 1627:42–95.Crossref, Google Scholar
• [30] Otto F, Villani C (2000) Generalization of an inequality by Talagrand and links with the logarithmic Sobolev inequality. J. Functional Anal. 173:361–400.Crossref, Google Scholar
• [31] Perchet V (2014) Approachability, regret and calibration: Implications and equivalences. J. Dynamic Games 1(2):181–254.Crossref, Google Scholar
• [32] Solan E, Vieille N (2002) Correlated equilibrium in stochastic games. Games Econom. Behav. 38:362–399.Crossref, Google Scholar
• [33] Sznitman AS (1991) Topics in propagation of chaos. École d’Été de Probabilités de Saint-Flour XIX. Lecture Notes Math. 1464:165–251.Crossref, Google Scholar