Cycles in adversarial regularized learning

MaRDI QIDQ4608067zbMATH OpenFDO

Authors Panayotis Mertikopoulos, Georgios Piliouras, Christos Papadimitriou

Publication date 15 March 2018

Full work available at URL https://arxiv.org/abs/1709.02738, http://dl.acm.org/citation.cfm?id=3175476

Mathematics Subject Classification ID

Entropy and other invariants, isomorphism, classification in ergodic theory (37A35) Learning and adaptive systems in artificial intelligence (68T05) Rationality and learning in game theory (91A26)

Recommendations

• Learning in games via reinforcement and regularization
• Replicator dynamics: old and new
• Learning, regret minimization, and equilibria
• Near-optimal no-regret algorithms for zero-sum games
• Q-learning in regularized mean-field games

Cited in

(37)

• Learning in games with continuous action sets and unknown payoff functions
• Physics-informed neural networks for gravity field modeling of the Earth and Moon
• Physics-informed neural networks for rarefied-gas dynamics: Poiseuille flow in the BGK approximation
• Exploration-exploitation in multi-agent learning: catastrophe theory meets game theory
• Understanding and Mitigating Gradient Flow Pathologies in Physics-Informed Neural Networks
• On gradient-based learning in continuous games
• A biologically plausible neural network for multichannel canonical correlation analysis
• scientific article; zbMATH DE number 7625188 (Why is no real title available?)
• Continuous-time convergence rates in potential and monotone games
• scientific article; zbMATH DE number 7064064 (Why is no real title available?)
• Zeroth-order algorithms for nonconvex-strongly-concave minimax problems with improved complexities
• A stochastic variant of replicator dynamics in zero-sum games and its invariant measures
• Riemannian Hamiltonian Methods for Min-Max Optimization on Manifolds
• A unified stochastic approximation framework for learning in games
• No-regret learning for repeated non-cooperative games with lossy bandits
• Fast convergence of optimistic gradient ascent in network zero-sum extensive form games
• Sig‐Wasserstein GANs for conditional time series generation
• Memory loss can prevent chaos in games dynamics
• Alleviating limit cycling in training GANs with an optimization technique
• Physics-informed neural networks and functional interpolation for stiff chemical kinetics
• Nested replicator dynamics, nested logit choice, and similarity-based learning
• On zero-sum game representation for replicator dynamics
• Mather sets of a cascade generated by the FTRL algorithm of a two-agent zero-sum game with mixed strategies
• Globally optimal strategy against the hedge algorithm in repeated games
• The emergence of chaos in population game dynamics induced by comparisons
• Unsynchronized decentralized Q-learning: two timescale analysis by persistence
• Fast computation of optimal transport via entropy-regularized extragradient methods
• Two-timescale gradient descent ascent algorithms for nonconvex minimax optimization
• On the discrete-time origins of the replicator dynamics: from convergence to instability and chaos
• The attractor of the replicator dynamic in zero-sum games
• The replicator dynamic, chain components and the response graph
• Last round convergence and no-dynamic regret in asymmetric repeated games
• Last-iterate convergence rates for min-max optimization: convergence of Hamiltonian gradient descent and consensus optimization
• Swim till you sink: computing the limit of a game
• O(1/T) time-average convergence in a generalization of network zero-sum games via alternating gradient descent
• Periodicity in Hedge-myopic system and an asymmetric NE-solving paradigm for two-player zero-sum games
• From Darwin to Poincaré and von Neumann: recurrence and cycles in evolutionary and algorithmic game theory

This page was built for publication: Cycles in adversarial regularized learning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4608067)