Deep Neural networks (DNNs), extensively applied across diverse disciplines, are characterized by their integrated and monolithic architectures, setting them apart from conventional software systems. This architectural difference introduces particular challenges to maintenance tasks, such as model restructuring (e.g., model compression), re-adaptation (e.g., fitting new samples), and incremental development (e.g., continual knowledge accumulation). Prior research addresses these challenges by identifying task-critical neuron layers, and dividing neural networks into semantically-similar sequential modules. However, such layer-level approaches fail to precisely identify and manipulate neuron-level semantic components, restricting their applicability to finer-grained model maintenance tasks. In this work, we implement NeuSemSlice, a novel framework that introduces the semantic slicing technique to effectively identify critical neuron-level semantic components in DNN models for semantic-aware model maintenance tasks. Specifically, semantic slicing identifies, categorizes and merges critical neurons across different categories and layers according to their semantic similarity, enabling their flexibility and effectiveness in the subsequent tasks. For semantic-aware model maintenance tasks, we provide a series of novel strategies based on semantic slicing to enhance NeuSemSlice. They include semantic components (i.e., critical neurons) preservation for model restructuring, critical neuron tuning for model re-adaptation, and non-critical neuron training for model incremental development. A thorough evaluation has demonstrated that NeuSemSlice significantly outperforms baselines in all three tasks.

翻译：深度神经网络（DNNs）已广泛应用于多个学科领域，其特点是采用集成式、整体化的架构，这使其有别于传统的软件系统。这种架构差异给模型维护任务带来了特殊挑战，例如模型重构（如模型压缩）、再适应（如适配新样本）和增量开发（如持续知识积累）。先前的研究通过识别任务关键神经元层，并将神经网络划分为语义相似的序列模块来应对这些挑战。然而，此类层级方法无法精确识别和操作神经元级语义组件，限制了其在更细粒度模型维护任务中的适用性。本工作实现了NeuSemSlice，这是一个新颖的框架，引入语义切片技术以有效识别DNN模型中关键的神经元级语义组件，用于语义感知的模型维护任务。具体而言，语义切片根据语义相似性识别、分类并合并不同类别和层中的关键神经元，从而确保其在后续任务中的灵活性和有效性。针对语义感知的模型维护任务，我们基于语义切片提出了一系列新颖策略以增强NeuSemSlice。这些策略包括：面向模型重构的语义组件（即关键神经元）保留、面向模型再适应的关键神经元微调，以及面向模型增量开发的非关键神经元训练。全面评估表明，NeuSemSlice在所有三项任务中均显著优于基线方法。