研究人员为卷积神经网络（CNN）开发了一种新颖的“在线架构”策略，显著增强了其翻译不变性。通过战略性地插入全局平均池化（GAP）层，该方法在保持ImageNet上具有竞争力的准确性的同时，将可训练参数减少了98%，网络规模减少了90%。该方法还提高了翻译鲁棒性，并已应用于感知图像质量评估，优于现有指标。 AI
影响 增强了CNN的鲁棒性和效率，可能改进图像分析和质量评估任务。
排序理由 详细介绍CNN新架构修改的学术论文。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2604.27870v1 Announce Type: new Abstract: Convolutional Neural Networks (CNNs) are widely assumed to be translation-invariant, yet standard architectures exhibit a startling fragility: even a single-pixel shift can drastically degrade performance due to their reliance on sp…
Convolutional Neural Networks (CNNs) are widely assumed to be translation-invariant, yet standard architectures exhibit a startling fragility: even a single-pixel shift can drastically degrade performance due to their reliance on spatially dependent fully connected layers. In thi…