Build notes · Goh Kun Ming
Residual Learning Study
Research · arXiv preprint · 2025
Project overview
Understanding why skip connections help deep networks.
A 2025 arXiv preprint co-authored with Xingyu Liu, studying residual connections through a ResNet-18 and conventional deep-CNN comparison on CIFAR-10.
Inside this build
- Studied the role of residual connections in gradient flow.
- Compared training behaviour and classification results on CIFAR-10.
- Documented the implementation and evaluation in a public preprint.
Project files
Original project files are in English.
-
Research preprint
Xingyu Liu, Kun Ming Goh arXiv:2510.24036v1 CC BY 4.0
Document preview
Use the page controls to browse. Enlarge a page to read the details.
Loading document…
When enlarged, scroll within the page to see more.
Page text
Text is extracted from the original English document. Charts, images and reading order may need the PDF for context.
