angular softmax
Type: Concept
As of November 2026, angular softmax has been mentioned in one podcast episode on Carnegie Mellon University Deep Learning (Jan 2026). Angular softmax is introduced as a margin-based loss that focuses on angles between feature vectors to improve class separation in zero-shot learning.
| Mentioned in | 1 episode on 1 podcast (as of November 2026) |
|---|---|
| First heard | Jan 2026 on Carnegie Mellon University Deep Learning |
| Field | Machine Learning |
Who talks about angular softmax on podcasts?
angular softmax came up; the most recent mention was Jan 2026 on Carnegie Mellon University Deep Learning.
- 0.20 Losses part1 on Carnegie Mellon University Deep Learning (Jan 2026)