-
### Links
- Paper : https://arxiv.org/abs/2006.09882
- Github : https://github.com/facebookresearch/swav
### 한 줄 요약
- SSL 분야에서 clustering 방식에 contrastive learning 방식을 적용한 논문으로, 기존의 clustering 방식…
-
1. Freeze teacher
2. add mlp on top
3. train teacher again with this new mlp (for CLIP just do contrastive learning for a bit)
4. distill this new teacher + MLP into student
-
![image](https://user-images.githubusercontent.com/46675408/150263175-a405612c-b130-4a53-ab18-88b8eb199107.png)
[arxiv](https://arxiv.org/pdf/2112.12750v1.pdf), [code](https://github.com/facebookrese…
-
Replacing contrastive learning from image level to pixel level,it's going to be a lot less efficient, especially when calculating comparative loss. So how do you solve the thorny efficiency issues?
-
We need to finish this very basic part of the code to get to the more interesting stuff later, i.e. the unsupervised classification/learning using contrastive divergence in restricted Boltzmann machin…
-
Collection of interesting papers we find interesting about EEG and SSL
-
I have tried a two-tower model (user and query) in a real industrial scenario using contrastive learning. The samples are all actual click samples, and the loss function is InfoNCE. I have a few quest…
-
Hello, I have read your "Contrastive Spatio-Temporal Representations for Reinforcement Learning" and related papers such as CURL. Can you provide your experimental code for my further study? Thanks.
-
How can I use GraphCL for fully unsupervised Graph Clustering?
So far, all that I've found the method for Graph Clustering is actually for node clustering or not a fully unsupervised learning metho…
-
Your work is very good and effective. But I have some questions about the baseline approach. I tried different hyperparameters to adjust supervised contrastivelearning or unsupervised contrastive lea…