SeqGAN

Sequence Generation Framework (SeqGAN) models the data generator as a stochastic policy in reinforcement learning (RL), SeqGAN bypasses the generator differentiation problem by directly performing gradient policy update.

References

Yu, L., Zhang, W., Wang, J., & Yu, Y. (2017). SeqGAN: Sequence Generative Adversarial Nets with Policy Gradient. ArXiv preprint ArXiv:1609.05473. Link