Optimal asset allocation for outperforming a stochastic benchmark target

We propose a data-driven Neural Network (NN) optimization framework to determine the optimal multi-period dynamic asset allocation strategy for outperforming general stochastic target. formulate problem as an control with asymmetric, distribution shaping, objective function. The proposed is illustrated in accumulation phase of defined contribution pension plan, goal achieving higher terminal wealth than benchmark. demonstrate that approach capable learning adaptive directly from historical market returns, without assuming any parametric model financial dynamics. outperforms benchmark constant proportion strategy, 90% probability, 46% median wealth, and significantly more right-skewed distribution.