TY - CHAP
T1 - Parabolic equations with quadratic growth in ℝn
AU - Bensoussan, Alain
AU - Frehse, Jens
AU - Peng, Shige
AU - Yam, Sheung Chi Phillip
PY - 2019
Y1 - 2019
N2 - We study here quasi-linear parabolic equations with quadratic growth in ℝn. These parabolic equations are at the core of the theory of PDE; see Friedman (Partial differential equations of parabolic type. Prentice-Hall, Englewood Cliffs, 1964) [6], Ladyzhenskaya et al. (Translations of Mathematical Monographs. AMS, 1968) [4] for details. However, for the applications to physics and mechanics, one deals mostly with boundary value problems. The boundary is often taken to be bounded and the solution is bounded. This brings an important simplification. On the other hand, stochastic control theory leads mostly to problems in ℝn. Moreover, the functions are unbounded and the Hamiltonian may have quadratic growth. There may be conflicts which prevent solutions to exist. In stochastic control theory, a very important development deals with BSDE (Backward Stochastic Differential Equations). There is a huge interaction with parabolic PDE in ℝn. This is why, although we do not deal with BSDE in this paper, we use many ideas from Briand and Hu (Probab Theory Relat Fields 141(3–4):543–567, 2008) [1], Da Lio and Ley (SIAM J Control Optim 45(1):74–106, 2006) [2], Karoui et al. (Backward stochastic differential equations and applications, Princeton BSDE Lecture Notes, 2009) [3], Kobylanski (Ann Probab 28(2):558–602, 2000) [5]. Our presentation provided here is slightly innovative.
AB - We study here quasi-linear parabolic equations with quadratic growth in ℝn. These parabolic equations are at the core of the theory of PDE; see Friedman (Partial differential equations of parabolic type. Prentice-Hall, Englewood Cliffs, 1964) [6], Ladyzhenskaya et al. (Translations of Mathematical Monographs. AMS, 1968) [4] for details. However, for the applications to physics and mechanics, one deals mostly with boundary value problems. The boundary is often taken to be bounded and the solution is bounded. This brings an important simplification. On the other hand, stochastic control theory leads mostly to problems in ℝn. Moreover, the functions are unbounded and the Hamiltonian may have quadratic growth. There may be conflicts which prevent solutions to exist. In stochastic control theory, a very important development deals with BSDE (Backward Stochastic Differential Equations). There is a huge interaction with parabolic PDE in ℝn. This is why, although we do not deal with BSDE in this paper, we use many ideas from Briand and Hu (Probab Theory Relat Fields 141(3–4):543–567, 2008) [1], Da Lio and Ley (SIAM J Control Optim 45(1):74–106, 2006) [2], Karoui et al. (Backward stochastic differential equations and applications, Princeton BSDE Lecture Notes, 2009) [3], Kobylanski (Ann Probab 28(2):558–602, 2000) [5]. Our presentation provided here is slightly innovative.
UR - http://www.scopus.com/inward/record.url?scp=85051219890&partnerID=8YFLogxK
UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85051219890&origin=recordpage
U2 - 10.1007/978-3-319-78325-3_8
DO - 10.1007/978-3-319-78325-3_8
M3 - RGC 12 - Chapter in an edited book (Author)
SN - 9783319783246
T3 - Computational Methods in Applied Sciences
SP - 91
EP - 110
BT - Contributions to Partial Differential Equations and Applications
A2 - Chetverushkin, B. N.
A2 - Fitzgibbon, W.
A2 - Kuznetsov, Y. A.
A2 - Neittaanmäki, P.
A2 - Periaux, J.
A2 - Periaux, O.
PB - Springer International Publishing
ER -