Posts Index
How to visualize hypergraphs with Python and networkx - The Easy Way
Graphs are awesome, hypergraphs are hyperawesome! Hypergraphs are a generalization of graphs where one relaxes the requirement for edges to connect just two nodes and allows instead edges to connect...
Should employees be promoted at random? Science says: “IDK, maybe?”
“In a hierarchy every employee tends to rise to his level of incompetence” – Dr. Laurence J. Peter
Borwein Integrals interpreted via Path Integrals
The ability to see common patterns and analogies, to conjecture universal laws starting from the particular things we can experience, is a powerful tool, which arguably is one of the...
Data science case studies: easy to avoid common mistakes
During my time as a Data Scientist, I had the chance to interview my fair share of candidates for data-related roles. While doing this, I started noticing a pattern: some...
Transfer learning made easy: let's build a dog breed classifier!
In this post, I want to showcase how simple it has become to perform transfer learning with the help of modern libraries. I will use TensorFlow, but the situation looks...
Time Series Forecasting with an LSTM Encoder/Decoder in TensorFlow 2.0
In this post I want to illustrate a problem I have been thinking about in time series forecasting, while simultaneously showing how to properly use some Tensorflow features which greatly...
On Differentiable Neural Architecture Search, DARTS, and Auto-Deeplab
I recently stumbled across a very interesting paper (thanks a lot, Noam!) titled “Auto-DeepLab: Hierarchical Neural Architecture Search for Semantic Image Segmentation”, by C. Liu et al.; the authors were...
Deep Learning, Heteroscedasticity, and TensorFlow Probability
It is very often the case that Machine Learning models calculate point estimates of quantities of interest; a typical example would be: given the number of rooms, size, and location...
