To address the aerodynamic load reduction requirement when the launch vehicle flying in high wind zone during the ascending phase, an intelligent attitude control method with adaptive learning rate is proposed. Taking a certain type of launch vehicle as the research object, the dynamic model in the pitch plane is established. A deep reinforcement learning framework suitable for flight control of the launch vehicle during the ascending phase is developed based on soft Actor-Critic (SAC), and a reward function that comprehensively considers control accuracy, control system stability, and load reduction effectiveness is designed. On this basis, an adaptive iteration of learning rate is implemented based on a step-size learning rate scheduler to quickly improve the convergence of the controller, and an early stopping mechanism which can automatically end the training process is designed to enhance training efficiency. Simulations show that the proposed method can effectively improve load reduction performance of the launch vehicle while ensuring attitude tracking accuracy and control system stability. Additionally, it has strong adaptability and generalization ability to external uncertain disturbances.