Integrating Offline Reinforcement Learning with Transformers for Sequential Recommendation

no code implementations • 26 Jul 2023 • Xumei Xi, Yuke Zhao, Quan Liu, Liwen Ouyang, Yang Wu

To this end, we train a farsighted recommender by using an offline RL algorithm with the policy network in our model architecture that has been initialized from a pre-trained transformer model.

Offline RL reinforcement-learning +1