PulseAugur
实时 17:55:59
English(EN) Miles: Metric Learning with Expandable Subspace for Pre-Trained Model-Based Class-Incremental Learning

新的Miles方法增强了预训练模型的类别增量学习能力

研究人员开发了一种名为Miles(可扩展子空间度量学习)的新方法,以改进预训练模型的类别增量学习(CIL)。现有的CIL方法要么遭受灾难性遗忘,要么通过为每个新任务扩展模型参数来增加计算成本。Miles通过引导优化有效地扩展参数空间来解决这个问题,利用预训练模型的先验知识。该方法解耦了可学习模块,并使用中间特征进行灵活扩展,在六个基准数据集上取得了最先进的性能。 AI

影响 这项研究可能导致更有效的方法来持续更新AI模型,而不会丢失先前学习到的信息。

排序理由 该集群描述了一篇关于类别增量学习新方法的最新研究论文。

在 Hugging Face Daily Papers 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

新的Miles方法增强了预训练模型的类别增量学习能力

报道来源 [2]

  1. Hugging Face Daily Papers TIER_1 English(EN) ·

    Miles: Metric Learning with Expandable Subspace for Pre-Trained Model-Based Class-Incremental Learning

    Class Incremental Learning (CIL) aims to learn new concepts consistently from a data stream without forgetting. Unlike typical CIL methods which need to learn a model from scratch, pre-trained model (PTM) can easily adapt to a new task with fine-tuning. However, existing PTM-base…

  2. arXiv cs.CV TIER_1 English(EN) · Kai Jiang, Zisong Lin, Hongyuan Zhang, Xueru Bai, Xuelong Li ·

    Miles:基于预训练模型的、可扩展子空间的度量学习用于类别增量学习

    arXiv:2607.17593v1 Announce Type: new Abstract: Class Incremental Learning (CIL) aims to learn new concepts consistently from a data stream without forgetting. Unlike typical CIL methods which need to learn a model from scratch, pre-trained model (PTM) can easily adapt to a new t…