https://www.chenk.top/en/tags/gradnorm/Recent content in GradNorm on Chen Kai BlogHugoenSat, 31 May 2025 09:00:00 +0000https://www.chenk.top/en/transfer-learning/06-multi-task-learning/Sat, 31 May 2025 09:00:00 +0000https://www.chenk.top/en/transfer-learning/06-multi-task-learning/<p>A self-driving car using a single camera needs to do three things simultaneously: detect cars and pedestrians, segment lanes and free space, and estimate the distance of each pixel. Training three separate networks would triple the parameters, require three times as many forward passes at inference, and overlook the fact that all three tasks need the same low-level features (edges, surfaces, occlusion cues).</p> <p>Multi-task learning (MTL) is the alternative: one shared backbone, one task-specific head per output, all trained jointly. Done well, you cut parameters by 60% <strong>and</strong> lift accuracy on every task because each task acts as a regularizer for the others. Done badly, two of your three tasks regress and you waste a week wondering why.</p>