实时可视化神经网络训练过程，帮助理解前向与反向传播
NeuroViz: Real-time Interactive Visualization of Forward and Backward Passes in Neural Network Training
- 交互式工具实时展示网络激活、权重更新和损失变化
- 用户研究显示易用性评分80.97，超70%参与者认为透明度显著提升
- 适合初学者理解训练机制，也适用于教学演示
神经网络训练过程难以解释，尤其对新手而言。我们提出NeuroViz，一个支持全连接神经网络训练实时探索的交互式可视化工具。用户可配置网络结构、激活函数、学习率和数据集，实时观察激活值、权重更新和损失变化。NeuroViz将权重变化与前向、反向传播中的激活信号直接对应，可在单个周期内区分更新前后状态，并动态展示每个神经元的计算方程。我们对31名参与者进行了对比用户研究，与六种现有可视化工具相比，获得最高可用性评分（SUS 80.97，属‘优秀’范围），清晰度均值2.47，实用性均值2.23（越低越好）。超过70%的参与者表示可视化显著提升了对训练过程的透明感知。系统已开源，访问地址：https://neuroviz.org。
原文摘要 · Abstract (English)
Training neural networks is difficult to interpret, particularly for newcomers. We introduce NeuroViz, an interactive visualization tool that supports real-time exploration of fully connected neural network training. Users can configure network architecture, activation functions, learning rates, and datasets, then observe activations, weight updates, and loss progression. NeuroViz visualizes weight changes in direct correspondence with activation signals in both forward and backward passes, enabling users to distinguish pre- and post-update states within individual epochs and view dynamically updating per-neuron equations. We conduct a comparative user study with 31 participants against six established visualization tools and we achieved the highest usability score (SUS 80.97, in the 'excellent' range), with mean rankings of 2.47 for clarity and 2.23 for usefulness (lower is better). Over 70% of participants reported that the visualizations substantially increased their perception of neural network training transparency. The implemented instance is accessible at https://neuroviz.org.
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。