Training Federated GANs with Theoretical Guarantees: A Universal Aggregation Approach | OpenReview

Training Federated GANs  with Theoretical Guarantees: A Universal Aggregation Approach | OpenReview

Recently, Generative Adversarial Networks (GANs) have demonstrated their potential in federated learning, i.e., learning a centralized model from data privately hosted by multiple sites. A...

1 mentions: @jellied_unagi
Keywords: gan
Date: 2021/01/13 02:21

Referring Tweets

@jellied_unagi t.co/Kw2pucVhKk でGANのoptimumの証明で「optimal discriminatorが得られている」という前提があることについて注文がついている(実際はoptimal discriminatorなんて手に入らないでしょ、みたいな)んだが、これを言い出すとGANの証明全部ダメになってしまうのでは

Related Entries

Read more [1910.10329] Is the Trotterized UCCSD Ansatz usefully well-defined?
0 users, 3 mentions 2019/10/24 05:19
Read more [2003.12687] Serialized Output Training for End-to-End Overlapped Speech Recognitioncontact arXivarX...
0 users, 0 mentions 2020/03/31 11:20
Read more GitHub - HuangYG123/CurricularFace: CurricularFace(CVPR2020)
0 users, 0 mentions 2020/04/14 11:21
Read more GitHub - KunalMGupta/NeuralMeshFlow: Neural Mesh Flow: 3D Manifold Mesh Generation via Diffeomorphic...
0 users, 0 mentions 2020/11/11 20:21
Read more NanoDet "Super fast and lightweight anchor-free object detection model. Real-time on mobile devices"...
0 users, 0 mentions 2020/11/29 03:51