Summary of Contrastive Learning Of Preferences with a Contextual Infonce Loss, by Timo Bertram et al.
Contrastive Learning of Preferences with a Contextual InfoNCE Loss
by Timo Bertram, Johannes Fürnkranz, Martin Müller
First submitted to arxiv on: 8 Jul 2024
Categories
- Main: Artificial Intelligence (cs.AI)
- 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 The paper addresses a common issue in contextual preference ranking by proposing an adapted version of the CLIP framework to solve the problem. The issue arises when comparing a single preferred action against multiple choices, leading to increased complexity and skewed preference distributions. The authors show how they can adapt the InfoNCE loss used by CLIP for computer vision and multi-modal domains to their specific use case. They empirically demonstrate the effectiveness of their adapted version in the domain of collectable card games, aiming to learn an embedding space that captures associations between single cards and whole pools based on human selections. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper helps solve a problem in ranking preferences by using a special kind of machine learning called CLIP. Right now, it’s hard to compare one preferred choice against many others because it makes things too complicated. The researchers found a way to make it work by adapting a technique used for computer vision and other domains. They tested this new method on collectable card games, where they wanted to learn how cards are related to each other based on what people like. |
Keywords
» Artificial intelligence » Embedding space » Machine learning » Multi modal
