Model Soups - Paper Explained
In this video, I want to present the key ideas of "Model soups: averaging weights of multiple fine-tuned models improves accuracy without increasing inference time". This paper introduces us to a new approach of fine-tuning models, which relieves the pain of determining the appropriate parameters and random seed.
📑 Chapters:
00:00 Introduction
00:54 Model Soup
01:32 Model Soup Recipes
01:58 Experiments
02:43 Intuition
03:28 Conclusion
📝 Link to the paper:
https://arxiv.org/abs/2203.05482
👥 Authors:
Mitchell Wortsman, Gabriel Ilharco, Samir Yitzhak Gadre, Rebecca Roelofs, Raphael Gontijo-Lopes, Ari S. Morcos, Hongseok Namkoong et al.
🔗 Helpful Links:
My Video on "CoAtNet: Marrying Convolution and Attention for All Data Sizes":
• CoAtNet: Marrying Convolution and Attentio...
🙋♂️ Find me on: halflingwizard.me
🎁 Support the Channel:
If you’d like to support my work, you can check out my wishlist here: https://throne.com/halflingwizard
Any contribution, big or small, is greatly appreciated and helps me continue creating content. Thank you so much for your support and for being part of this journey!
#deep_learning #machine_learning #ai #computer_vision