Stochastic approximation algorithms and applications

By C. Kubrusly, J. Gravier. 1973

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Variational Inference: A Review for Statisticians (2016, methodology): “Because it rests on optimization, variational inference easily takes advantage of methods like stochastic optimization (Robbins and Monro 1951; Kushner and Yin 1997) and distributed optimization.”
A Tutorial on Bayesian Optimization of Expensive Cost Functions, with Application to Active User Modeling and Hierarchical Reinforcement Learning (2010, background): “Stochastic approximation is a popular idea for optimizing unknown objective functions in machine learning contexts [Kushner and Yin, 1997].”
Decentralized RLS With Data-Adaptive Censoring for Regressions Over Large-Scale Networks (2016, unknown): “The merits of RLS algorithms in solving centralized linear regression problems are well recognized [11], [25].”
Joint Scheduling of URLLC and eMBB Traffic in 5G Wireless Networks (2017, methodology): “To conclude the proof, Lemmas 2 and 3 along with the condition 3 satisfy all the conditions necessary to apply Theorem 2.1 in Chapter 5, [22] which states that R ( t ) converges to r ∗ almost surely.”
Locally Stationary Processes (2011, background): “…methods” (cf. Ljung (1977); Ljung and Söderström (1983)), in the stochastic approximation community (cf. Benveniste, Métivier and Priouret (1990); Kushner and Yin (1997)), in the neural network community under the name “back-propagation algorithm” (cf. White (1992) or by Haykin (1994)), and in…”
Comparing Value-Function Estimation Algorithms in Undiscounted Problems (2012, methodology): “If convergence for noisy processes is required then one can compute the averages of the estimated values in a second step { this method is called Poljak-averaging and has been considered by a number of authors [5, 4].”
A framework for opportunistic scheduling in wireless networks (2003, methodology): “We refer readers to [10,21] for a systematic and rigorous study of stochastic approximation algorithms.”
Autonomous reinforcement learning with experience replay. (2013, background): “It is the main principle of a wide class of stochastic approximation algorithms (Kushner & Yin, 1997).”
Distributed learning and cooperative control for multi-agent systems (2009, background): “Our theoretical results are based on the ordinary differential equation (ODE) approach [17] [18].”
Varieties of learning automata: an overview (2002, methodology): “It is similar to stochastic approximation algorithms [34] though here the randomness in choosing the next parameter value makes the algorithm explore better search directions.”
Online learning control by association and reinforcement (2000, unknown): “mization problems [10].”
A review of simulation optimization techniques (1998, background): “the reader is referred to L’Ecuyer (1991), Fu (1994), Fu and Hu (1997), and references therein.”
Bayesian inference of chemical kinetic models from proposed reactions (2015, methodology): “An effective strategy is to use a stochastic approximation method (Kushner and Yin, 1997; Robbins and Monro, 1951) that couples posterior exploration with the solution of the minimization problem.”
Escaping Nash Inflation (2000, methodology): “We want to think about these data in terms of a model with the following features: (1) the monetary authority controls the inflation rate, apart from a random disturbance; (2) the true data generating mechanism embodies a version of the natural rate hypothesis embedded within an expectational Philli”
Opportunistic transmission scheduling with resource-sharing constraints in wireless networks (2001, background): “The first bar is the normalized performance value under the ideal condition; the second bar is that of estimated thresholds with ideal measurements (i.e., the exact value of is known).”
Monte Carlo sampling-based methods for stochastic optimization (2014, background): “A different line of work involves the use of ordinary differential equations to analyze the behavior of SA algorithms; such an approach was introduced by Kushner and Clark [139] and elaborated further in Kushner and Yin [140], see also Borkar and Meyn [31].”
Design of Experiments in Nonlinear Models: Asymptotic Normality, Optimality Criteria and Small-Sample Properties (2013, background): “For θ̂ the LS estimator, Lai (1994) suggests the following design sequence: ⎧ ⎨ ⎩ xk = x ∗ 1(θ̂ k−1) if k is even and k ∈ {k1, k2 .”