计算社会科学“论文工作坊”2019暑期课

20190701

1. Understanding neural networks with Pytorch
2. Intro of word embeddings

20190702

1. word2vec examples with sequential numbers
2. discussing research topics

• Douban research, Zhengyi Liang

20190703

1. RNN & CNN
2. discussing research topics

• The shape of story, Huimin Xu
• cultrual dimension
• Using svd method finding the main diension.
• deepwalk, network embeddings.

20190704

1. CNN
2. discussing the applications of deep learning methods & eci method of hidalgo's research.
3. HITS and ECI

20190705

1. transE
2. RNN

Readings

day 1

• The book of Deep Learning
  • Chapter 1
• 几种梯度下降方法介绍
• 如何在网络很差的情况下安装pytorch？

day 2

• Mikolov et al. (2013). Efficient Estimation of Word Representations in Vector Space
• Mikolov et al. (2013). Distributed representations of words and phrases and their compositionality

day 3

• Semantics derived automatically from language corpora contain human-like biases. Caliskan, A., Bryson, J. J. and Narayanan, A. (2017). Science, 356 (6334). pp. 183-186. ISSN 0036-8075
• The Geometry of Culture: Analyzing Meaning through Class through Word Embeddings. Austin C. Kozlowski, Matt Taddy, James A. Evans. AJS.
• Word embeddings quantify 100 years of gender and ethnic stereotypes, Nikhil Garga, Londa Schiebinger, Dan Jurafsky, and James Zou. 2017. Pnas.
• The product space conditions the development of nations. Hidago, 2007, Science.
• The book of Deep Learning, Chapter 2. svd method

day 4

• The building blocks of economic complexity
• Effects of Exposure to Political Protests on Political Discussion and Attitudes in Authoritarian Regimes: Evidence from Eight Natural Experiments with Chinese Social Media, Zhang Han, AJS.
• CASM: A Deep Learning Approach for Identifying Collective Action Events with Text and Image Data from Social Media. Han Zhang∗ Jennifer Pan. 2019.

day 5

• Unsupervised word embeddings capture latent knowledge from materials science literature