torch.autograd.grad, PyTorch Developers, 2024 (PyTorch Foundation) - Official PyTorch documentation detailing the functional interface for computing gradients, including how to enable higher-order differentiation with create_graph=True.
Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks, Chelsea Finn, Pieter Abbeel, and Sergey Levine, 2017Proceedings of the 34th International Conference on Machine Learning, Vol. 70 (PMLR) - Introduces Model-Agnostic Meta-Learning (MAML), a meta-learning algorithm that relies on differentiating through optimization steps, thus requiring higher-order gradients.
Improved Training of Wasserstein GANs, Ishaan Gulrajani, Faruk Ahmed, Martin Arjovsky, Vincent Dumoulin, Aaron Courville, 2017Advances in Neural Information Processing Systems, Vol. 30DOI: 10.48550/arXiv.1704.00028 - Presents Wasserstein GANs with Gradient Penalty (WGAN-GP), a technique that stabilizes GAN training by enforcing a Lipschitz constraint via a gradient penalty, which is computed using higher-order differentiation.