Long-Horizon Q-Learning: Accurate Value Learning via n-Step Inequalities

Paper Link: https://arxiv.org/pdf/2605.05812