New work on the arxiv: "Network reconstruction and community detection from dynamics",
https://arxiv.org/abs/1903.10833
I show how coupling Bayesian network reconstruction from functional behavior with community detection enhances both tasks simultaneously.
Just out on PRX: "Reconstructing Networks with Unknown and Heterogeneous Errors"
New method can reconstruct networks and provide error estimates for them, even when measurement uncertainties are unknown.
https://journals.aps.org/prx/abstract/10.1103/PhysRevX.8.041011
Code is available as part of graph-tool: https://graph-tool.skewed.de
The documentation for the reconstruction code is here: https://graph-tool.skewed.de/static/doc/demos/inference/inference.html#network-reconstruction
Each integer is represented in a high-dimensional space, and gets squished down to 2D so that numbers with similar prime factorisations are closer together than those with dissimilar factorisations.
https://johnhw.github.io/umap_primes/index.md.html
Via @svscarpino@twitter.com.
"Science starts with a question. Data science starts with the data. What makes data science so hard is that it starts in the wrong place."
Great analogy of scientific protocol with law & order episodes!
Now I can't get the title song out of my head.
https://simplystatistics.org/2018/08/15/the-law-and-order-of-data-science/
Finally on the arXiv: "Reconstructing networks with unknown and heterogeneous errors" https://arxiv.org/abs/1806.07956
Did you know you can reconstruct and make error estimates for networks, by making only a single noisy measurement?
Physics & Networks