机器学习人工学weekly-2018/5/27

1. Google相关

1.1 DeepMind最新meta learning论文

Prefrontal cortex as a meta-reinforcement learning system

链接：https://deepmind.com/blog/prefrontal-cortex-meta-reinforcement-learning-system/?utm_campaign=ARCHITECHT&utm_medium=email&utm_source=ARCHITECHT_28

1.2 Google Clips相机如何抓拍精彩瞬间

Automatic Photography with Google Clips

链接：https://ai.googleblog.com/2018/05/automatic-photography-with-google-clips.html

1.3 机器学习最佳实践

Rules of Machine Learning: Best Practices for ML Engineering

链接：https://developers.google.com/machine-learning/rules-of-ml/?utm_campaign=Revue%20newsletter&utm_medium=Newsletter&utm_source=Deep%20Learning%20Weekly

2. 腾讯用RL和MCTS打王者荣耀的论文来了

Feedback-Based Tree Search for Reinforcement Learning

链接：https://arxiv.org/pdf/1805.05935.pdf

3. Auto-ML相关

3.1 auto-sklearn

链接：https://github.com/automl/auto-sklearn

3.2 TPOT

链接：https://github.com/EpistasisLab/tpot

4. tensor编译器对比

Tensor Compilers: Comparing PlaidML, Tensor Comprehensions, and TVM

链接：http://vertex.ai/blog/compiler-comparison

5. RL和进化算法混搭解决货运船搬集装箱问题

Evolutionary RL for Container Loading

链接：https://arxiv.org/abs/1805.06664

6. Airbnb用机器学习给房间图片分类

Categorizing Listing Photos at Airbnb

链接：https://medium.com/airbnb-engineering/categorizing-listing-photos-at-airbnb-f9483f3ab7e3

7. 康奈尔大学的数据科学基础(书)

链接：https://www.cs.cornell.edu/jeh/book.pdf

8. RL在知识图谱reasoning领域的应用

8.1 Go for a Walk and Arrive at the Answer: Reasoning Over Paths in Knowledge Bases using Reinforcement Learning

链接：https://arxiv.org/abs/1711.05851

8.2 DeepPath: A Reinforcement Learning Method for Knowledge Graph Reasoning

链接：https://arxiv.org/abs/1707.06690

9. 用机器学习做adblocker

AdGraph: A Machine Learning Approach to Automatic and Effective Adblocking

链接：https://arxiv.org/pdf/1805.09155.pdf

10. 教程

10.1 Monte Carlo Tree Search教程

A Deep Dive into Monte Carlo Tree Search

链接：http://www.moderndescartes.com/essays/deep_dive_mcts/

10.2 easy-tensorflow教程

链接：https://github.com/easy-tensorflow/easy-tensorflow?utm_campaign=Data_Elixir&utm_medium=email&utm_source=Data_Elixir_183