Hard to understand for a newbie in Deep RL. A formal definition of what is a skill ("A skill is simply a policy.") would help. ### Typos: - "guaranteeing that is has maximum entropy" - "We discuss the the log p(z) term in Appendix B." - "so it much first gather momentum" - "While are skills are learned"
I agree with the previous comment. The article seems to aim at people who are already familiar with reinforcement (not necessarily deep, or based on neural networks I guess) and its usual benchmark. The implementations are not detailed, the authors lay stress on the general idea (which is relatively simple to get), and its visual results which look quite spectacular.
> For example, an author would not add a link to a Wikipedia page, as it is considered (rightly or wrongly) under the scientific quality standards I would not agree, some scientists already write blog-posts about their beloved subjects using a lot of hyperlinks to Wikipedia and other sites, consider for example [Igor Pak's](https://igorpak.wordpress.com/2015/05/26/the-power-of-negative-thinking-part-i-pattern-avoidance/) or [Terence Tao's](https://terrytao.wordpress.com/2009/04/26/szemeredis-regularity-lemma-via-random-partitions/) blogs. Currently, such blog-posts are considered complementary to "real" scientific publications. But but one day everything will change.

### General: A very simple and very scalable algorithm with theoretical guarantees to compute an approximation of the closeness centrality for each node in the graph. Maybe the framework can be adapted to solve other problems. In particular, "Hoeffding's theorem" is very handy. ### Parallelism: Note that the algorithm is embarrassingly parallel (the loop over the $k$ reference nodes can be done in parallel with almost no additional work). ### Experiments: It would be nice to have some experiments on large real-world graphs. An efficient (and parallel) implementation of the algorithm is available here: https://github.com/maxdan94/BFS ### Related work: More recent related work by Paolo Boldi and Sebastiano Vigna: - http://mmds-data.org/presentations/2016/s-vigna.pdf - https://arxiv.org/abs/1011.5599 - https://papers-gamma.link/paper/37 It also gives an approximation of the centrality for each node in the graph and it also scales to huge graphs. Which algorithm is "the best"?