https://www.chenk.top/en/tags/metric-learning/Recent content in Metric Learning on Chen Kai BlogHugoenMon, 19 May 2025 09:00:00 +0000https://www.chenk.top/en/transfer-learning/04-few-shot-learning/Mon, 19 May 2025 09:00:00 +0000https://www.chenk.top/en/transfer-learning/04-few-shot-learning/<p>Show a child one photograph of a pangolin and they will spot pangolins for life. Show a deep learning model one photograph and it will give you a uniformly random guess. Few-shot learning is the field that closes that gap: building classifiers that work with only one to ten labeled examples per class.</p> <p>The trick is not to memorize individual classes harder. It is to learn <em>how to learn</em> from very few examples, then carry that ability over to brand-new classes at test time. This article covers the two families that dominate the field today: <strong>metric learning</strong>, which learns a good distance function, and <strong>meta-learning</strong>, which learns a good initialization.</p>