TY - JOUR AB - Singular perturbation techniques allow the derivation of an aggregate model whose solution is asymptotically optimal for Markov decision processes with strong and weak interactions. We develop an algorithm that takes advantage of the asymptotic optimality of the aggregate model in order to compute the solution of the original model. We derive conditions for which the proposed algorithm has better worst case complexity than conventional contraction algorithms. Based on our complexity analysis, we show that the major benefit of aggregation is that the reduced order model is no longer ill conditioned. The reduction in the number of states (due to aggregation) is a secondary benefit. This is a surprising result since intuition would suggest that the reduced order model can be solved more efficiently because it has fewer states. However, we show that this is not necessarily the case. Our theoretical analysis and numerical experiments show that the proposed algorithm can compute the optimal solution with a reduction in computational complexity and without any penalty in accuracy. AU - Ho,CP AU - Parpas,P DO - 10.1137/130944254 EP - 3886 PY - 2014/// SN - 0363-0129 SP - 3854 TI - Singularly perturbed markov decision processes: a multiresolution algorithm T2 - SIAM Journal on Control and Optimization UR - http://dx.doi.org/10.1137/130944254 UR - http://gateway.webofknowledge.com/gateway/Gateway.cgi?GWVersion=2&SrcApp=PARTNER_APP&SrcAuth=LinksAMR&KeyUT=WOS:000346845100017&DestLinkType=FullRecord&DestApp=ALL_WOS&UsrCustomerID=1ba7043ffcc86c417c072aa74d649202 UR - https://epubs.siam.org/doi/10.1137/130944254 UR - http://hdl.handle.net/10044/1/21116 VL - 52 ER -
Transition to Zero Pollution is a flagship initiative of the Imperial's Academic Strategy, with a vision to realise a sustainable zero pollution future. The initiative brings researchers from different disciplines together to take a systems approach to tackling pollution in all its forms.