- Implementation of
Zhou, R., Wang, X., Zhang, N., Lv, X., & Li-Jia, L. (2017). Deep Reinforcement Learning-based Image Captioning with Embedding Reward. ArXiv.org, Apr 12, 2017. - We used Advantage Actor Critic (A2C) Model to train the Policy network and the value network.
- Also experimented with curriculum learning-based approaches.
- Project completed as part of Reinforcement Learning class at SJSU, Spring 2020
- Presentation
