Kappen, Hilbert J.; Gómez, Vicenç; Opper, Manfred
(Springer, 2012)
We reformulate a class of non-linear stochastic optimal control problems introduced by Todorov (in Advances in Neural Information Processing Systems, vol. 19, pp. 1369-1376, 2007) as a Kullback-Leibler (KL) minimization ...