-
Hi! This is more of a little excursion than a 'true' issue, but it's about a technique which I've found useful and would like to share. The occasion is that I'm extending my library [zimt](https://git…
-
Have you tried using deconvolution instead of upsampling?
-
Hi @JingweiJ Thanks for the wonderful work. Do you plan to release the baseline models for the proposed tasks, i.e. (few-shot) action recognition, spatial-temporal scene graph prediction? That would g…
-
**Is your feature request related to a problem? Please describe.**
A relatively common text classification task I encounter is hierarchical multi-label classification, like classifying scientific pap…
-
Since we plan to use PyPy for running time experiments in our paper, we should check that the findings that Willem made in CPython are roughly reflected in PyPy, in particular that when Powersort subs…
-
**Describe the feature**
Features description
**Motivation**
A clear and concise description of the motivation of the feature. Ex1. It is inconvenient when [....]. Ex2. There is a recent paper [.…
-
Hi @karpathy,
We built [papers.labml.ai](https://papers.labml.ai) in May ([introductory tweet](https://twitter.com/labmlai/status/1395978352175902726)) to discover research papers based on populari…
-
I would just like to share my experience after working with the `ppo2` PPO implementation. I happily welcome objections, in case I just haven't found the right way to work with the implementations.
…
-
I want to modify the conv and deconv pipeline based on this concept [conv_arithmetic](https://github.com/vdumoulin/conv_arithmetic) rather than my previous one. My previous deconvolution treat forward…
-
- [ ] Normalized Adaptive Gradient
```
@article{DBLP:journals/corr/abs-1305-6646,
author = {St{\'{e}}phane Ross and
Paul Mineiro and
John Langford},
title…