机器学习人工学weekly-2018/5/20

注意下面很多链接需要科学上网，无奈国情如此

1. TF条件控制实现

Dynamic Control Flow in Large-Scale Machine Learning

链接：https://arxiv.org/abs/1805.01772

2. 用RL玩Chrome里面里面的恐龙游戏

TF-rex: Playing Google's T-rex game with TensorFlow

链接：https://vdutor.github.io/blog/2018/05/07/TF-rex.html

3. fast.ai 2018年part 2上线了

Launching Cutting Edge Deep Learning for Coders: 2018 edition

链接：http://www.fast.ai/2018/05/07/part2-launch/

4. GluonCV也是做model zoo

GluonCV — Deep Learning Toolkit for Computer Vision

链接：https://medium.com/apache-mxnet/gluoncv-deep-learning-toolkit-for-computer-vision-9218a907e8da

5. Gmail Smart Composer

Smart Compose: Using Neural Networks to Help Write Emails

链接：https://ai.googleblog.com/2018/05/smart-compose-using-neural-networks-to.html

6. OpenAI对算力的总结

AI and Compute

链接：https://blog.openai.com/ai-and-compute/?

7. RNN鲁棒性的研究

How Robust are Deep Neural Networks?

链接：https://arxiv.org/abs/1804.11313?utm_campaign=Revue%20newsletter&utm_medium=Newsletter&utm_source=Deep%20Learning%20Weekly

8. 数据增强（使用fastai库）

Data Augmentation using Fastai

链接：https://becominghuman.ai/data-augmentation-using-fastai-aefa88ca03f1

9. 推荐系统简介

Introduction to Recommender Systems in 2018

链接：https://tryolabs.com/blog/introduction-to-recommender-systems/