学习资源

1. Crash Cousrse 公开课,天文地理

MIT Technology Review Reading Notes

1. GAN (generative adversarial network) can forge faces, arts, or any images, which means they have great potential to be used in ethically objectionalble and dangerous way, such as criminal, overlay celebrity faces.

2. Climate changes could affect the cost of company because the overheated empoyees may be too hot to work.

3. Researchers from Stanford university used machine learning model to identify the signals of depression, which learned from facial expression, voice tone, and spoken words.

4. Quantum computers are on their way. Their computing power can be used for cryptography. Decryption of old encrypted data is a big threat against security and privacy. Encryption standards needs to be developed in the next several decades for quantum-proof cryptographic algorithms to be widely adopted.

自学总结 self-study summary

1. Build GAN-based classifier using Keras

2. Find that precision, recall, f1, roc, and auc API from sklearn maybe the most convenient way, compared to Keras’s metrics callback.

小句子 Inspiration

1. from movie: The Matrix 黑客帝国. 那么我为什么在做现在手上的事？

Because you didn’t come here to make the choice. You’ve already made it. You’re here to try to understand why you made it.

你来到这里不是为了做选择，你早已选择了。你来到这里的目的，是为了了解你为什么这样选择