Summary of One Step Learning, One Step Review, by Xiaolong Huang et al.
One Step Learning, One Step Review
by Xiaolong Huang, Qiankun Li, Xueran Li, Xuesong Gao
First submitted to arxiv on: 19 Jan 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Machine Learning (cs.LG)
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 This paper proposes a novel approach to visual fine-tuning, called OLOR (One step Learning, One step Review), which combines fine-tuning with optimizers and incorporates a weight rollback term to mitigate knowledge forgetting. The method ensures consistency in the weight range of upstream and downstream models, enhancing fine-tuning performance. A layer-wise penalty is also presented to adjust weight rollback levels for adapting varying downstream tasks. Experimental results on image classification, object detection, semantic segmentation, and instance segmentation demonstrate the general applicability and state-of-the-art performance of OLOR. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps solve a problem with how we make computer vision models learn new skills. Right now, when we fine-tune these models for a specific task, they tend to forget what they learned before. The researchers propose a new way to do this fine-tuning that keeps the model’s old knowledge and makes it perform better overall. |
Keywords
* Artificial intelligence * Fine tuning * Image classification * Instance segmentation * Object detection * Semantic segmentation
