We tackle the problem of novel class discovery, which aims to learn novel classes without supervision based on labeled data from known classes. A key challenge lies in transferring the knowledge in the known-class data to the learning of novel classes. Previous methods mainly focus on building a shared representation space for knowledge transfer and often ignore modeling class relations. To address this, we introduce a class relation representation for the novel classes based on the predicted class distribution of a model trained on known classes. Empirically, we find that such class relation becomes less informative during typical discovery training. To prevent such information loss, we propose a novel knowledge distillation framework, which utilizes our class-relation representation to regularize the learning of novel classes. In addition, to enable a flexible knowledge distillation scheme for each data point in novel classes, we develop a learnable weighting function for the regularization, which adaptively promotes knowledge transfer based on the semantic similarity between the novel and known classes. To validate the effectiveness and generalization of our method, we conduct extensive experiments on multiple benchmarks, including CIFAR100, Stanford Cars, CUB, and FGVC-Aircraft datasets. Our results demonstrate that the proposed method outperforms the previous state-of-the-art methods by a significant margin on almost all benchmarks. Code is available at \href{https://github.com/kleinzcy/Cr-KD-NCD}{here}.

翻译：我们解决了新类别发现问题，其目标是在已知类别的标注数据基础上，无需监督地学习新类别。关键挑战在于将已知类别数据中的知识迁移至新类别的学习过程。以往方法主要侧重于构建共享表征空间以实现知识迁移，但通常忽略了对类关系的建模。为此，我们基于已知类别训练模型的预测类别分布，为新类别引入了一种类关系表征。实验发现，在典型的新类别发现训练过程中，该类关系的信息量会逐渐衰减。为避免这种信息损失，我们提出了一种新颖的知识蒸馏框架，利用所提出的类关系表征来正则化新类别的学习过程。此外，为使新类别中每个数据点能灵活适配知识蒸馏方案，我们设计了可学习的权重函数用于正则化，该函数基于新类别与已知类别的语义相似度自适应地促进知识迁移。为验证方法的有效性与泛化能力，我们在包括CIFAR100、Stanford Cars、CUB和FGVC-Aircraft数据集在内的多个基准上进行了广泛实验。结果表明，所提方法在几乎所有基准上均以显著优势超越以往的先进方法。代码已发布于\href{https://github.com/kleinzcy/Cr-KD-NCD}{此处}。