Oscillatory exchange bias and training effects in nanocrystalline Pr0.5Ca0.5MnO3

We report on exchange bias effects in 10 nm particles of Pr0.5Ca0.5MnO3 which appear as a result of competing interactions between the ferromagnetic (FM)/anti-ferromagnetic (AFM) phases. The fascinating new observation is the demonstration of the temperature dependence of oscillatory exchange bias (OEB) and is tunable as a function of cooling field strength below the SG phase, may be attributable to the presence of charge/spin density wave (CDW/SDW) in the AFM core of PCMO10. The pronounced training effect is noticed at 5 K from the variation of the EB field as a function of number of field cycles (n) upon the field cooling (FC) process. For n > 1, power-law behavior describes the experimental data well; however, the breakdown of spin configuration model is noticed at n >= 1. Copyright 2012 Author(s). This article is distributed under a Creative Commons Attribution 3.0 Unported License. http://dx.doi.org/10.1063/1.3696033]

An Online Actor-Critic Algorithm with Function Approximation for Constrained Markov Decision Processes

We develop an online actor-critic reinforcement learning algorithm with function approximation for a problem of control under inequality constraints. We consider the long-run average cost Markov decision process (MDP) framework in which both the objective and the constraint functions are suitable policy-dependent long-run averages of certain sample path functions. The Lagrange multiplier method is used to handle the inequality constraints. We prove the asymptotic almost sure convergence of our algorithm to a locally optimal solution. We also provide the results of numerical experiments on a problem of routing in a multi-stage queueing network with constraints on long-run average queue lengths. We observe that our algorithm exhibits good performance on this setting and converges to a feasible point.