Hacker News
- How to Train 10k-Layer Vanilla Convolutional Neural Networks [pdf] https://proceedings.mlr.press/v80/xiao18a/xiao18a.pdf 3 comments
- New Research shows it's possible to generate plausible Medical Imaging counterfactuals using Generative AI & Causality (see demo in the comments) https://proceedings.mlr.press/v202/de-sousa-ribeiro23a.html 2 comments science
- [D] Why weren't Normalized Advantage Functions more widely adopted? http://proceedings.mlr.press/v48/gu16.pdf 6 comments reinforcementlearning
- 3D rendering as a differentiable function function http://proceedings.mlr.press/v80/athalye18b.html?ref=https%3A%2F%2Fgithubhelp.com 3 comments graphicsprogramming
- Unsupervised RL: ICM and the nosiy TV problem https://proceedings.mlr.press/v70/pathak17a.html 3 comments reinforcementlearning
- [D] Do CNN representations really correlate with the pixel space? http://proceedings.mlr.press/v97/goyal19a/goyal19a.pdf 8 comments machinelearning
- Leaking private data from gradients [D] https://proceedings.mlr.press/v54/mcmahan17a/mcmahan17a.pdf 3 comments machinelearning
- What are "well-behaved" gradients? http://proceedings.mlr.press/v37/jozefowicz15.pdf 3 comments learnmachinelearning