【新智元导读】2026年架构革命的枪声已经打响!ResNet用十年证明了「加法捷径」能救训练,但也暴露了「只加不减」的天花板。DeepSeek新年王炸之后,普林斯顿和UCLA新作DDL让网络学会忘记、重写和反转。
作为计算机视觉领域的顶尖科学家,何恺明毕业于清华大学物理系,现任麻省理工学院电气工程与计算机科学系副教授(终身教职)。其最著名的成就,就是提出了ResNet架构,解决了深度神经网络中“越深越难训练”的根本性问题,ResNet也被称为现代深度学习发展史 ...
The paper comes at a time when most AI start-ups have been focusing on turning AI capabilities in LLMs into agents and other ...
Deep Learning with Yacine on MSNOpinion

What are 1x1 convolutions in deep learning – explained simply

Understand how 1x1 convolutions work and why they’re essential in modern neural network architectures like ResNet and ...