A Reinforcement Learning-Based Control Approach for Unknown Nonlinear Systems with Persistent Adversarial Inputs | IEEE Conference Publication | IEEE Xplore