search

grypesc / SEED

ICLR2024 paper on Continual Learning
https://arxiv.org/abs/2401.10191
MIT License
28 stars 5 forks source link

About the parameters in Table 3 #1

Closed BlacknoSheep closed 6 months ago

BlacknoSheep commented 6 months ago

In Table 3, it is shown that SEED utilizes 3.2 million parameters, whereas ResNet18 has 11.7 million parameters. Could you clarify which part of the parameters is referred to by "#Params." in the table?

grypesc commented 6 months ago

Hello, thanks for interest in this work. As stated in Appendix A.1, we utilize ResNet32 for Tab.3 which has ~0.46M parameters. Then, you can calculate the number of total parameters (network weights and Gausses) using Eq. 3 in Appendix A.3. Hope that helps!

BlacknoSheep commented 6 months ago

Hello, thanks for interest in this work. As stated in Appendix A.1, we utilize ResNet32 for Tab.3 which has ~0.46M parameters. Then, you can calculate the number of total parameters (network weights and Gausses) using Eq. 3 in Appendix A.3. Hope that helps!

Thanks!