Effective exploration is key to a successful search process.The recently proposed negatively correlated search(NCS) tries to achieve this by coordinated parallel exploration,where a set of search processes are driven to be negatively cor-related so that different promising areas of the search space can be visited simultaneously.Despite successful applications of NCS,the negatively correlated search behaviors were mostly devised by intuition,while deeper (e.g.,mathematical) under-standing is missing.In this paper,a more principled NCS,namely NCNES,is presented,showing that the parallel explo-ration is equivalent to a process of seeking probabilistic models that both lead to solutions of high quality and are distant from previous obtained probabilistic models.Reinforcement learn-ing,for which exploration is of particular importance,are con-sidered for empirical assessment.The proposed NCNES is ap-plied to directly train a deep convolution network with 1.7 mil-lion connection weights for playing Atari games.Empirical re-sults show that the significant advantages of NCNES,especially on games with uncertain and delayed rewards,can be highly owed to the effective parallel exploration ability.