无需预设网络宽度，可生成跨架构的完整神经网络。
NNiT: Width-Agnostic Neural Network Generation with Structurally Aligned Weight Spaces
- 将权重矩阵分块为局部结构化字段，实现宽度无关生成。
- 在未见架构上达85%以上成功率，显著优于基线方法。
- 适合需要跨架构泛化的神经网络生成场景。
神经网络参数的生成常受限于具体架构，因标准参数表示依赖已知的权重矩阵维度。此外，排列对称性使得不同参数化网络仍能实现相似输入输出功能，进一步增加生成难度。本文提出神经网络扩散变换器（NNiT），通过将权重矩阵分块为补丁并建模为局部结构场，实现宽度无关的权重生成。我们证明，结合卷积神经网络解码器的图超网络（GHN）能结构性对齐权重空间，建立补丁处理所需的局部相关性。以多层感知机（MLP）为例，尽管其排列对称性明显，NNiT仍可在多种架构上生成完整功能性网络。该方法在单序列模型中联合建模离散架构标记与连续权重补丁。在ManiSkill3机器人任务中，对训练时未见过的拓扑结构，NNiT成功率达85%以上，而基线方法无法泛化；同一框架亦可推广至MNIST分类任务，超越原始控制场景。
原文摘要 · Abstract (English)
Generative modeling of neural network parameters is often tied to architectures because standard parameter representations rely on known weight-matrix dimensions. Generation is further complicated by permutation symmetries that allow networks to model similar input-output functions while having widely different, unaligned parameterizations. In this work, we introduce Neural Network Diffusion Transformers (NNiTs), which generate weights in a width-agnostic manner by tokenizing weight matrices into patches and modeling them as locally structured fields. We establish that Graph HyperNetworks (GHNs) with a convolutional neural network (CNN) decoder structurally align the weight space, creating the local correlation necessary for patch-based processing. Focusing on Multilayer Perceptrons (MLPs), where permutation symmetry is especially apparent, NNiTs generate fully functional networks across a range of architectures. Our approach jointly models discrete architecture tokens and continuous weight patches within a single sequence model. On ManiSkill3 robotics tasks, NNiT achieves $>85\%$ success on architecture topologies unseen during training, while baseline approaches fail to generalize; the same pipeline also generalizes to MNIST classification beyond the robotic control setting.
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。