Summary of Boosting Meta-training with Base Class Information For Few-shot Learning, by Weihao Jiang et al.
Boosting Meta-Training with Base Class Information for Few-Shot Learning
by Weihao Jiang, Guodong Liu, Di He, Kun He
First submitted to arxiv on: 6 Mar 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Computer Vision and Pattern Recognition (cs.CV)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed end-to-end training paradigm for few-shot learning consists of two alternative loops. The outer loop calculates cross entropy loss on the entire training set while updating only the final linear layer. In the inner loop, the original meta-learning training mode is employed to calculate the loss and incorporate gradients from the outer loss to guide parameter updates. This approach not only converges quickly but also outperforms existing baselines, indicating that information from the overall training set and the meta-learning training paradigm could mutually reinforce one another. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Few-shot learning helps computers recognize new things with just a few examples. Meta-learning is a way to achieve this. A recent method called Meta-Baseline does well, but it’s not the best because it takes too long and doesn’t use all the data. To fix this, researchers created an end-to-end training method that uses two loops. The outer loop looks at the whole dataset while updating only one part of the model. The inner loop uses meta-learning to improve the model. This approach works better than others and is flexible, allowing it to work with different models. |
Keywords
* Artificial intelligence * Cross entropy * Few shot * Meta learning
