https://www.chenk.top/en/tags/adversarial-learning/Recent content in Adversarial Learning on Chen Kai BlogHugoenTue, 13 May 2025 09:00:00 +0000https://www.chenk.top/en/transfer-learning/03-domain-adaptation/Tue, 13 May 2025 09:00:00 +0000https://www.chenk.top/en/transfer-learning/03-domain-adaptation/<p>Your autonomous-driving stack works perfectly on sunny California freeways. Then it rains in Seattle. Top-1 accuracy drops from 95% to 70%. The model did not get worse — the <em>data distribution shifted</em>, and your training set never told it what wet asphalt looks like at dusk.</p> <p>This is the everyday problem of <strong>domain adaptation</strong>: you have abundant labelled data in one distribution (the <em>source</em>) and unlabelled data in another (the <em>target</em>), and you need the model to perform on the target. This article shows you how, from first-principles theory to a working DANN implementation.</p>