https://terms-en.ai-term-hub.com/zh/tags/ml-paradigm/Recent content in ML Paradigm on 中文AI术语词典Hugozh-cnSat, 18 Jul 2026 11:44:45 +0000https://terms-en.ai-term-hub.com/zh/terms/multiple_instance_learning/Sat, 18 Jul 2026 11:01:16 +0000https://terms-en.ai-term-hub.com/zh/terms/multiple_instance_learning/<h2 id="definition">Definition</h2> <p>多示例学习（MIL）解决的是数据被分组为具有单一标签的“包”，而包内的各个实例未标记的场景。通常，如果一个包中包含至少一个正例实例，则该包被标记为正例；否则为负例。</p> <h3 id="summary">Summary</h3> <p>一种弱监督学习范式，标签被分配给实例集合（包），而非单个实例。</p> <h2 id="key-concepts">Key Concepts</h2> <ul> <li>包级标注</li> <li>实例级不确定性</li> <li>弱监督</li> <li>正/负包逻辑</li> </ul> <h2 id="use-cases">Use Cases</h2> <ul> <li>药物活性预测</li> <li>基于边界框的图像分类</li> <li>基于内容的图像检索</li> </ul> <h2 id="related-terms">Related Terms</h2> <ul> <li><a href="https://terms-en.ai-term-hub.com/en/terms/weak_supervision-%E5%BC%B1%E7%9B%91%E7%9D%A3/">weak_supervision (弱监督)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/bagging-%E8%A3%85%E8%A2%8B%E6%B3%95/">bagging (装袋法)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/instance_classification-%E5%AE%9E%E4%BE%8B%E5%88%86%E7%B1%BB/">instance_classification (实例分类)</a></li> <li><a href="https://terms-en.ai-term-hub.com/en/terms/label_noise-%E6%A0%87%E7%AD%BE%E5%99%AA%E5%A3%B0/">label_noise (标签噪声)</a></li> </ul>