Techniques for feedforward networks (FFNs) and convolutional networks (CNNs) are frequently reused across families, but the relationship between the underlying model classes is rarely made explicit. We introduce a unified node-level formalization with tensor-valued activations and show that generalized feedforward networks form a strict subset of generalized convolutional networks. Motivated by the mismatch in per-input parameterization between the two families, we propose model projection, a parameter-efficient transfer learning method for CNNs that freezes pretrained per-input-channel filters and learns a single scalar gate for each (output channel, input channel) contribution. Projection keeps all convolutional layers adaptable to downstream tasks while substantially reducing the number of trained parameters in convolutional layers. We prove that projected nodes take the generalized FFN form, enabling projected CNNs to inherit feedforward techniques that do not rely on homogeneous layer inputs. Experiments across multiple ImageNet-pretrained backbones and several downstream image classification datasets show that model projection is a strong transfer learning baseline under simple training recipes.

翻译：前馈网络（FFNs）与卷积网络（CNNs）的技术常在不同网络家族间复用，但两类基础模型类别之间的关联却鲜少被明确阐述。本文引入一种具有张量值激活函数的统一节点级形式化描述，并证明广义前馈网络构成广义卷积网络的严格子集。针对两类网络在每输入参数化方式上的不匹配问题，我们提出模型投影——一种面向卷积网络的参数高效迁移学习方法，该方法冻结预训练的每输入通道滤波器，并为每个（输出通道，输入通道）贡献学习单个标量门控。投影机制使所有卷积层保持对下游任务的适应能力，同时显著减少卷积层中需训练的参数数量。我们证明投影后的节点呈现广义前馈网络形式，从而使投影后的卷积网络能够继承不依赖同质层输入的前馈网络技术。在多个基于ImageNet预训练的主干网络及若干下游图像分类数据集上的实验表明，模型投影在简单训练方案下可作为强大的迁移学习基线方法。