Stochastic Optimal Control

By Alekos Cecchin, Markus Fischer. 1989

How researchers see this paper

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

In Reinforcement Learning

Citations in the corpus

Way Off-Policy Batch Deep Reinforcement Learning of Implicit Human Preferences in Dialog (2019, methodology): “The approach we propose is based on KL-control, a branch of stochastic optimal control (SOC) [58] where the Kullback-Leibler (KL) divergence from some distribution is used to regularize an RL policy (e.g. [1, 33, 48, 61]).”
Sequence Tutor: Conservative Fine-Tuning of Sequence Generation Models with KL-control (2016, methodology): “, 2012) is a branch of stochastic optimal control (SOC) (Stengel, 1986), which studies an RL, or control, problem in which the agent tries maximizing its task reward while minimizing deviation from a prior policy.”
Reinforcement Learning: A Survey (1996, neutral): “Adaptive control (Burghes & Graham, 1980; Stengel, 1986) is also concerned with algorithms for improving a sequence of decisions from experience.”
Locally Weighted Learning for Control (1997, unknown): “following linear dynamics: x(t+ 1) = A x(t) +Bu(t) (16) The optimal action with respect to the criteria in Equation 14 and linear dynamics in Equation 16 can be obtained by solution of a matrix equation called the Ricatti equation (Stengel, 1986).”
Robot trajectory optimization using approximate inference (2009, unknown): “As mentioned in the introduction, the LQG case is classically introduced as a perturbation model around an optimal trajectory of the deterministic system (Bryson & Ho, 1969; Stengel, 1986).”
Tuning Recurrent Neural Networks with Reinforcement Learning (2016, methodology): “1, our approach is related to stochastic optimal control (SOC) (Stengel, 1986) and KL control (Todorov, 2006; Kappen et al.”
Capacities, Measurable Selection and Dynamic Programming Part II: Application in Stochastic Control Problems (2013, methodology): “(i) Notice that with u.s.a. is upper semi-analytic (see e.g. [2, Corollary 7.48]).”
Stochastic control for a class of nonlinear kernels and applications (2015, unknown): “This is exactly the approach followed by Bertsekas and Shreve [6], and Dellacherie [24] for discrete time stochastic control problems.”
Optimal dividend strategies for a risk process under force of interest (2008, unknown): “We would like to thank Hanspeter Schmidli for stimulating discussions on the topic.”
Constructing Sublinear Expectations on Path Space (2012, background): “Let C ⊆ [1, 2] × [1, 2] be the graph of f , and define the random variable”
Risk-averse dynamic programming for Markov decision processes (2010, unknown): “Several excellent textbooks and monographs [5,6,21,27,32,39,50] discuss various aspects of this vast area.”
Learning to Walk via Deep Reinforcement Learning (2018, methodology): “This framework has been used in many contexts, from inverse reinforcement learning [46] to optimal control [43, 44, 36].”
Capacities, Measurable Selection and Dynamic Programming Part I: Abstract Framework (2013, methodology): “In particular, this framework can be considered as a continuous time extension of the discrete time dynamic programming of Bertsekas and Shreve [1], or the gambling house model studied in Dellacherie [15], Maitra and Sudderth [25].”