This paper presents a fast ICA algorithm that works best under Gaussian noise. This is demonstrated with components simulated from different univariate distributions and variable Gaussian noise. 

The writing is clear. The paper is incremental in the sense that it builds on ideas from (Belkin et. al, 2013) but focuses on speeding up and improving their cumulant-based approach. 
This is achieved via 
1. a Hessian expansion of the cumulant-tensor-based quasi-orthogonalization. 
2. gradient-based iterations that preserve quasi-orthogonalization of the latent factors (noised case) as well as whitening in the noiseless case. 

This paper proposes a cumulant based independent component analysis (ICA) algorithm for source separation in the presence of additive Gaussian noise. The algorithm is somewhat incremental building upon Refs [2] and [3], but appear technically correct with experimental results confirming the claims made. The algorithms used for benchmarking assume no additive noise but is like InfoMax often quite robust to addition of noise. 

This paper takes up the question of whether rhetorical relations can be automatically derived and classified. It focuses, in particular, on discourse markers. These may be ambigious (e.g 'since', 'yet' have multiple uses and are sometimes, but not always, discourse markers); and these discourse markers may also be missing altogether.

The authors comment that: "what is needed is a model which can classify rhetorical relations in the absence of an explicit discourse marker." (p4). Previous work (e.g. Marcu & Echihabi 2002) has suggested creating training data for a classifier by labelling examples which contain an unambiguous lexically marked rhetorical relation, then removing the markers. The main purpose of this paper is to empirically test this.

It also provides an interesting theoretical observation: Two conditions are needed for training on marked examples to work well:

"First, there has to be a certain amount of redundancy between the discourse marker and the general linguistic context, i.e. removing the discourse marker should still leave enough residual information for the classifier to learn how to distinguish different relations."

Second, similarity between marked and unmarked examples is needed so that a classifier can make generalizations.

The paper suggests that texts with lexically marked and lexically unmarked rhetorical relations may be inherently different, in so far as removing discourse markers may change the meaning of a sentence, and classifiers built based on removing markers from classified sentences work little better than chance.

- Implementations:    
    - https://hub.docker.com/r/mklinov/caffe-flownet2/
    - https://github.com/lmb-freiburg/flownet2-docker
    - https://github.com/lmb-freiburg/flownet2
- Explanations:
    - A Brief Review of FlowNet - not a clear explanation
    https://medium.com/towards-data-science/a-brief-review-of-flownet-dca6bd574de0
    - https://www.youtube.com/watch?v=JSzUdVBmQP4
Supplementary material: 
http://openaccess.thecvf.com/content_cvpr_2017/supplemental/Ilg_FlowNet_2.0_Evolution_2017_CVPR_supplemental.pdf

A comprehensive and accessible recent review detailing the methodological advances in the assessment of the gut microbiota

In the Neotropics, most plants depend on animals for pollination. Solitary bees are the most important vectors, and among them members of the tribe Centridini depend on oil from flowers (mainly Malpighiaceae) to feed their larvae. This specialized relationship within 'the smallest of all worlds' (a whole pollination network) could result in a 'tiny world' different from the whole system. This 'tiny world' would have higher nestedness, shorter path lengths, lower modularity and higher resilience if compared with the whole pollination network. In the present study, we contrasted a network of oil-flowers and their visitors from a Brazilian steppe ('caatinga') to whole pollination networks from all over the world. A network approach was used to measure network structure and, finally, to test fragility. The oil-flower network studied was more nested (NODF = 0·84, N = 0·96) than all of the whole pollination networks studied. Average path lengths in the two-mode network were shorter (one node, both for bee and plant one-mode network projections) and modularity was lower (M = 0·22 and four modules) than in all of the whole pollination networks. Extinctions had no or small effects on the network structure, with an average change in nestedness smaller than 2% in most of the cases studied; and only two species caused coextinctions. The higher the degree of the removed species, the stronger the effect and the higher the probability of a decrease in nestedness. We conclude that the oil-flower subweb is more cohesive and resilient than whole pollination networks. Therefore, the Malpighiaceae have a robust pollination service in the Neotropics. Our findings reinforce the hypothesis that each ecological service is in fact a mosaic of different subservices with a hierarchical structure ('webs within webs').

Theoretical and practical relevance: This paper goes one step further in the hypothesis of mutualistic modules, evidencing that ecosystem services may be a mosaic of subservices with different properties. Furthermore, this finding has important implications for service-oriented conservation programs, as their planning should take into account this hierarchical structure.