Summary of Can Contrastive Learning Refine Embeddings, by Lihui Liu and Jinha Kim and Vidit Bansal
Can Contrastive Learning Refine Embeddings
by Lihui Liu, Jinha Kim, Vidit Bansal
First submitted to arxiv on: 11 Apr 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: None
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 introduces a novel contrastive learning framework called SIMSKIP, which refines input embeddings for downstream tasks using output embeddings from previously trained encoder models. Unlike traditional unsupervised learning approaches, SIMSKIP takes advantage of these output embeddings as its input. The authors provide theoretical evidence that applying SIMSKIP does not result in larger upper bounds on downstream task errors than those of the original embeddings. Experimental results demonstrate that the embeddings produced by SIMSKIP improve performance on various open datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper creates a new way to learn representation using contrastive learning. It uses old encoder models’ outputs as inputs, which makes it different from traditional unsupervised learning methods. The authors show that this method doesn’t make things worse and actually helps with downstream tasks like image classification or language translation. They tested it on some datasets and found that it works well. |
Keywords
* Artificial intelligence * Encoder * Image classification * Translation * Unsupervised
