Bag of Tricks in Machine Learning

Attention for interpretability:

• Learning to Explain: https://arxiv.org/abs/1802.07814  
• Attention is not Explanation: https://arxiv.org/abs/1902.10186  
• Is Attention Interpretable?: https://arxiv.org/abs/1906.03731

Reviews:

• https://lilianweng.github.io/lil-log/2018/06/24/attention-attention.html 
• https://arxiv.org/abs/1902.02181 
• https://arxiv.org/abs/1904.02874 
• https://arxiv.org/abs/1811.05544 

• Why Did I Reject a Data Scientist Job?
• Why You Don’t Need Data Scientists
• Here’s why so many data scientists are leaving their jobs
• Why are Machine Learning Projects so Hard to Manage?
• Data Scientists: Why are they so expensive to hire?

Glad to appear in my friend Rand Xie’s talk! See: 22:29

Thanks for mentioning me. See 17:10

https://ai.googleblog.com/2019/04/evaluating-unsupervised-learning-of.html

The choice of random seed across different runs has a larger impact on disentanglement scores than the model choice and the strength of regularization (while naively one might expect that more regularization should always lead to more disentanglement). A good run with a bad hyperparameter can easily beat a bad run with a good hyperparameter.