[인공지능 논문 Review - 02]Meta-learning with warped gradients
1. Juho Lee, Yoonho Lee, Jungtaek Kim, Adam R. Kosiorek, Seungjin Choi, Yee Whye Teh (2019),
Proceedings of the Thirty-Sixth International Conference on Machine Learning (ICML-2019),
Long Beach, California, USA, June 9-15, 2019.
(earlier version in preprint arXiv:1810.00825 )
2. Juho Lee, Lancelot James, Seungjin Choi, and François Caron (2019),
Proceedings of the 22nd International Conference on Artificial Intelligence and Statistics (AISTATS-2019),
Naha, Okinawa, Japan, April 16-18, 2019. (oral)
(earlier version in preprint arXiv:1810.01778 )
3. Yoonho Lee and Seungjin Choi (2018),
in Proceedings of the Thirty-Fifth International Conference on Machine Learning (ICML-2018),
Stockholm, Sweden, July 10-15, 2018.
(earlier version in preprint arXiv:1810.05558 )
4. Saehoon Kim, Jungtaek Kim, and Seungjin Choi (2018),
in Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence (AAAI-2018),
New Orleans, Louisiana, USA, February 2-7, 2018.
5. Juho Lee, Creighton Heaukulani, Zoubin Ghahramani, Lancelot James, and Seungjin Choi (2017),
in Proceedings of the International Conference on Machine Learning (ICML-2017),
Sydney, Australia, August 6-11, 2017.
(earlier version in preprint arXiv:1702.08239 )
6. Saehoon Kim and Seungjin Choi (2017),
in Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence (AAAI-2017),
San Francisco, California, USA, February 4-9, 2017.
[Meta-learning with warped gradients]
In my previous posting, I introduced the MT-net [Lee and Choi, 2018] for gradient-based meta-learning, where T-layers (which are meta-learned) are interleaved between the layers of a task-specific learner.
Such an idea was generalized, leading to an excellent work,
‘warped gradient descent’, to be presented in ICLR-2020.
Nonlinear warp-layers are introduced to replace T-layers, the parameterized preconditioning matrix is efficiently meta-learned by the warped gradient descent.
A few things that I thought about in the MT-net now become clear,
thanks to the nice work by [Flennerhag et al, 2020].
논문보기 링크↓
https://arxiv.org/pdf/1909.00025.pdf