-
### Deep Learning Simplified Repository (Proposing new issue)
:red_circle: **Project Title** : Classification of Elon Musk Tweets using NLP
:red_circle: **Aim** : Create a classification model using…
-
## Neural Architecture Search with Reinforcement Learning
- 第一篇比较正式的 RL-Based NAS paper
- RL
- Agent - RNN - Controller - `A variable-length String`
- Reward - Accuracy
- Policy Gradient
…
-
Post your questions here about: “[Training and Taming Deep Networks](https://docs.google.com/document/d/1gne-oWcJs1p5sEjUumapq6HKeaOet3EHxJ-Ij0LuTro/edit?usp=sharing)” & “[The Expanding Universe of De…
-
TTT
-
Another paper EVALUATING THE SEARCH PHASE OF NEURAL ARCHITECTURE SEARCH tested FairNAS on NASBench101 but get the Kendall Tau of -0.23.
FairNAS using 13 models to evaulate the rank and get the Ken…
-
- Add examples for how to use the module.
- Must add examples for both, Jupyter notebook and Python file.
- Examples for standard BO include:
- [ ] Hyperparameter tuning for an ML model (like SVM…
-
It's always going to be possible to try new model architectures and look for better performance in this way. This task is more limited in it's ambition.
Our current Q architecture is based on sugge…
-
https://102.alibaba.com/fund/proposalTopicDetail.htm?id=1116
Alibaba Fund 主题
-
Greetings, Chao,
Your recent works on using CNN to enhance MoHex 2.0 and to guide proof search are really interesting! Solving Go hasn't made progress for nearly 10 years, and I wonder if we can do…
-
This net is the current state-of-the-art on ImageNet and CIFAR-10: _GPipe: Efficient Training of Giant Neural Networks using Pipeline Parallelism_ https://arxiv.org/abs/1811.06965
Would be great to…