@inproceedings{neuraltemporaldifferencelearningconverge, title = {Neural Temporal-Difference and Q-Learning Provably Converge to Global Optima}, author = {Qi Cai and Zhuoran Yang and Jason D. Lee and Zhaoran Wang}, year = {2019}, booktitle = {NeurIPS 2019 12}, eprint = {1905.10027}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/1905.10027v2}, }