Nonconvergence to unstable points in urn models and stochastic approximations
From MaRDI portal
Publication:921715
DOI10.1214/aop/1176990853zbMath0709.60054OpenAlexW2015836240MaRDI QIDQ921715
Publication date: 1990
Published in: The Annals of Probability (Search for Journal in Brave)
Full work available at URL: https://doi.org/10.1214/aop/1176990853
Related Items (68)
When can the two-armed bandit algorithm be trusted? ⋮ Generalized urn models of evolutionary processes. ⋮ Vertex reinforced random walks with exponential interaction on complete graphs ⋮ Multiscale Q-learning with linear function approximation ⋮ Vertex-reinforced random walks and a conjecture of Pemantle ⋮ Asymptotic pseudotrajectories and chain recurrent flows, with applications ⋮ Distributed dynamic reinforcement of efficient outcomes in multiagent coordination and network formation ⋮ Escaping Strict Saddle Points of the Moreau Envelope in Nonsmooth Optimization ⋮ Two repelling random walks on \(\mathbb{Z}\) ⋮ Phase transitions in non-linear urns with interacting types ⋮ Global convergence of the gradient method for functions definable in o-minimal structures ⋮ Risk-Sensitive Reinforcement Learning via Policy Gradient Search ⋮ Convergence of the Momentum Method for Semialgebraic Functions with Locally Lipschitz Gradients ⋮ How Fast Is the Bandit? ⋮ Heterogeneous beliefs and local information in stochastic fictitious play ⋮ First-order methods almost always avoid strict saddle points ⋮ Sufficient Conditions for Instability of the Subgradient Method with Constant Step Size ⋮ Learning in monotone Bayesian games ⋮ Interacting nonlinear reinforced stochastic processes: Synchronization or non-synchronization ⋮ Stochastic heavy ball ⋮ Dynamics of Morse-Smale urn processes ⋮ Two-timescale stochastic gradient descent in continuous time with applications to joint online parameter estimation and optimal sensor placement ⋮ Coexistence in Preferential Attachment Networks ⋮ A time-dependent version of Pólya's urn ⋮ Stochastic learning in multi-agent optimization: communication and payoff-based approaches ⋮ A Newton-Based Method for Nonconvex Optimization with Fast Evasion of Saddle Points ⋮ Second-Order Guarantees of Distributed Gradient Algorithms ⋮ Convergent multiple-timescales reinforcement learning algorithms in normal form games ⋮ Global Convergence of Policy Gradient Methods to (Almost) Locally Optimal Policies ⋮ Attracting edge and strongly edge reinforced walks ⋮ Vertex-reinforced random walk ⋮ Stochastic optimization with momentum: convergence, fluctuations, and traps avoidance ⋮ Negatively reinforced balanced urn schemes ⋮ Nonconvergence to saddle boundary points under perturbed reinforcement learning ⋮ On a notion of partially conditionally identically distributed sequences ⋮ Quasi-Newton smoothed functional algorithms for unconstrained and constrained simulation optimization ⋮ Stochastic approximation, cooperative dynamics and supermodular games ⋮ NEWTONIAN MECHANICS AND NASH PLAY ⋮ Learning, information, and sorting in market entry games: theory and evidence ⋮ A behavioral learning process in games ⋮ Unnamed Item ⋮ Nonlinear randomized urn models: a stochastic approximation viewpoint ⋮ On the convergence of reinforcement learning ⋮ Avoidance of traps in stochastic approximation ⋮ The actor-critic algorithm as multi-time-scale stochastic approximation. ⋮ Stochastic approximation algorithms: overview and recent trends. ⋮ Urn models and differential algebraic equations ⋮ Learning to signal: Analysis of a micro-level reinforcement model ⋮ Mixed equilibria and dynamical systems arising from fictitious play in perturbed games ⋮ A Herding Perspective on Global Games and Multiplicity ⋮ Learning across games ⋮ Adaptive dynamics in games played by heterogeneous populations ⋮ An Adjusted Payoff-Based Procedure for Normal Form Games ⋮ Mutation, Sexual Reproduction and Survival in Dynamic Environments ⋮ Gradient Descent Only Converges to Minimizers: Non-Isolated Critical Points and Invariant Regions ⋮ An ODE method to prove the geometric convergence of adaptive stochastic algorithms ⋮ DRAWING MULTISETS OF BALLS FROM TENABLE BALANCED LINEAR URNS ⋮ On generalized Pólya urn models ⋮ On the fast convergence of random perturbations of the gradient flow ⋮ Natural actor-critic algorithms ⋮ Self-interacting diffusions. III: Symmetric interactions ⋮ Time to absorption in discounted reinforcement models. ⋮ Attainability of boundary points under reinforcement learning ⋮ One-dimensional system arising in stochastic gradient descent ⋮ Forbidden Transactions and Black Markets ⋮ On (un)knots and dynamics in games ⋮ Extending the Step-Size Restriction for Gradient Descent to Avoid Strict Saddle Points ⋮ On Gradient-Based Learning in Continuous Games
This page was built for publication: Nonconvergence to unstable points in urn models and stochastic approximations
