Issue #6

July 7, 2019

---

1. 1. How Artificial Intelligence Is Changing Science [quantamagazine.org]
2. 2. Matlab vs. Julia vs. Python [tobydriscoll.net]
3. 3. The Thousand Brains Theory of Intelligence [youtube.com]
4. 4. Advanced Data Structures [courses.csail.mit.edu]
5. 5. Why do Neural Networks Need an Activation Function? [datastuff.tech]
6. 6. Queueing theory [github.com/joelparkerhenderso]
7. 7. How A.I. Helped Improve Crowd Counting in Hong Kong Protests [nytimes.com]

---

• • Unsupervised word embeddings capture latent knowledge from materials science literature (V. Tshitoyan, J. Dagdelen, L. Weston, A. Dunn, Z. Rong, O. Kononova, K. A. Persson, G. Ceder, A. Jain)
• • Machine Reading Comprehension: a Literature Review (X. Zhang, A. Yang, S. Li, Y. Wang)
• • Time Series Anomaly Detection with Variational Autoencoders (C. Zhang, Y. Chen)
• • Towards Assigning Priorities in Queues Using Age of Information (J. Xu, N. Gautam)
• • Social evolution leads to persistent corruption (J-H Lee, Y. Iwasa, U. Dieckmann, K. Sigmund)
• • Testing Heaps' law for cities using administrative and gridded population data sets (F. Simini, C. James)
• • Multivariate Spatiotemporal Hawkes Processes and Network Reconstruction (B. Yuan, H. Li, A. L. Bertozzi, P. J. Brantingham, M. A. Porter)

---