思想与社会2018 · 分享于 Hacker News
Measuring the Tendency of CNNs to Learn Surface Statistical Regularities
探索《Measuring the Tendency of CNNs to Learn Surface Statistical Regularities》的原文与 Hacker News 社区讨论,查看来源并加入阅读清单。
从论文,走向讨论
看看社区如何讨论这个想法。
来自 Hacker News 的原始讨论。不同的视角、疑问,以及值得继续思考的线索。
最多展示 100 条相关讨论。论文可能出现在评论中,并非原帖投稿内容。
在讨论中发现的引用
按 HN 条目顺序展示最近的 50 条引用来源。
YtheDoug ·
Cloud AutoML: Making AI accessible to every business
191 赞同93 评论
Ysytelus ·
Fourier transforms of images (2017)
174 赞同36 评论
Ymathattack ·
Why do CNNs generalize so poorly to small image transformations?
149 赞同46 评论
Yfanf2 ·
Deep learning: a critical appraisal
147 赞同61 评论
Ytosh ·
Why neural networks struggle with the Game of Life
141 赞同68 评论
Yrusht ·
Improving Deep Learning Performance with AutoAugment
132 赞同14 评论
Ypaufernandez ·
Adversarial Reprogramming of Neural Networks
98 赞同16 评论
Ytim_sw ·
Measuring the Tendency of CNNs to Learn Surface Statistical Regularities
66 赞同36 评论
YHarj ·
Bayesian Inference for Hiring Engineers
50 赞同60 评论
Ybad_ramen_soup ·
Ask HN: If another AI winter is upon us, where are the research chokepoints?
4 赞同6 评论
Ynafizh ·
CNNs learn surface statistical regularities, not high-level abstract concepts
2 赞同0 评论