用生成伪样本实现二值化嵌入的增量学习，兼顾性能与内存效率
Generative Binary Memory: Pseudo-Replay Class-Incremental Learning on Binarized Embeddings
- 基于伯努利混合模型生成二值化伪示例，建模类别分布多模态特性
- 在CIFAR100和TinyImageNet上分别提升2.9%和1.5%平均准确率
- 适用于嵌入式系统，支持二值神经网络，内存减少4.7倍
在新概念持续出现的动态环境中，深度神经网络需在不遗忘旧知识的前提下学习新类别，这正是类别增量学习（CIL）的核心挑战。本文提出生成式二值记忆（GBM），一种新型的CIL伪重放方法，通过生成合成的二值伪示例来缓解灾难性遗忘。该方法基于伯努利混合模型（BMM），在潜在二值空间中有效建模类别的多模态分布特征。结合专门设计的特征二值化器，该方法可适配任意传统DNN，并天然支持二值神经网络（BNN），适用于资源受限的嵌入式系统。实验表明，搭载该二值化器的ResNet-18在CIFAR100上平均准确率提升2.9%，在TinyImageNet上提升1.5%。在CORE50数据集上，相比新兴的CIL方法，最终准确率提升3.1%，同时内存占用降低4.7倍。
原文摘要 · Abstract (English)
In dynamic environments where new concepts continuously emerge, Deep Neural Networks (DNNs) must adapt by learning new classes while retaining previously acquired ones. This challenge is addressed by Class-Incremental Learning (CIL). This paper introduces Generative Binary Memory (GBM), a novel CIL pseudo-replay approach which generates synthetic binary pseudo-exemplars. Relying on Bernoulli Mixture Models (BMMs), GBM effectively models the multi-modal characteristics of class distributions, in a latent, binary space. With a specifically-designed feature binarizer, our approach applies to any conventional DNN. GBM also natively supports Binary Neural Networks (BNNs) for highly-constrained model sizes in embedded systems. The experimental results demonstrate that GBM achieves higher than state-of-the-art average accuracy on CIFAR100 (+2.9%) and TinyImageNet (+1.5%) for a ResNet-18 equipped with our binarizer. GBM also outperforms emerging CIL methods for BNNs, with +3.1% in final accuracy and x4.7 memory reduction, on CORE50.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。