Summary of Attraction-repulsion Swarming: a Generalized Framework Of T-sne Via Force Normalization and Tunable Interactions, by Jingcheng Lu et al.
Attraction-Repulsion Swarming: A Generalized Framework of t-SNE via Force Normalization and Tunable Interactions
by Jingcheng Lu, Jeff Calder
First submitted to arxiv on: 15 Nov 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI); Classical Analysis and ODEs (math.CA); Dynamical Systems (math.DS); Numerical Analysis (math.NA); Machine Learning (stat.ML)
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 method, ARS visualization, is a new approach to data visualization based on attraction-repulsion swarming (ARS) dynamics. Building upon t-distributed stochastic neighbor embedding (t-SNE), this framework views the visualization technique as a swarm of interacting agents driven by attraction and repulsion. The modified dynamics incorporate total influence normalization, allowing for a fixed time step and simple iteration without requiring optimization tricks. This approach also enables separate tuning of attraction and repulsion kernels, giving users control over cluster tightness and inter-cluster spacing in visualizations. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary ARS visualization is a new way to see data that uses ideas from swarms. Instead of using special tricks like t-SNE does, this method lets you make good choices about how the data looks. It’s better because it doesn’t need those tricks. You can also control how close things are within groups and far apart between them. |
Keywords
* Artificial intelligence * Embedding * Optimization
