Summary of Robust Semi-supervised Learning For Self-learning Open-world Classes, by Wenjuan Xi et al.
Robust Semi-Supervised Learning for Self-learning Open-World Classes
by Wenjuan Xi, Xin Song, Weili Guo, Yang Yang
First submitted to arxiv on: 15 Jan 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 proposes an open-world semi-supervised learning (SSL) method called Self-learning Open-world Classes (SSOC), which can accurately classify known classes and fine-grained distinguish different unknown classes. Existing SSL methods assume labeled and unlabeled data share the same class space, but in real-world applications, this assumption may not hold. SSOC addresses this challenge by defining class center tokens for both known and unknown classes, autonomously learning token representations using cross-attention mechanism, and designing a pairwise similarity loss to exploit information from instances’ predictions and relationships. The proposed method outperforms state-of-the-art baselines on multiple popular classification benchmarks. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps us learn better with just some labeled data and lots of unlabeled data. Imagine trying to teach a computer to recognize pictures, but only giving it a few examples of what’s in each picture, while having thousands more that it hasn’t seen before. The computer needs to figure out how to use the little information we gave it to also learn about all those new pictures. This paper shows a way to do just that using something called Self-learning Open-world Classes (SSOC). It works by identifying patterns in both the labeled and unlabeled data, which helps the computer learn even better. |
Keywords
* Artificial intelligence * Classification * Cross attention * Semi supervised * Token
