derivt

Sorry, I am not a native English speaker, but I think it should be White Asian, do you say white americans?

Also, Google accused in lawsuit of excluding white and Asian men in hiring to boost diversity

https://www.usatoday.com/story/tech/news/2018/03/01/google-a...

Higher education is not overestimated if it gives you better opportunities for the future. Perhaps some kind of higher education provides little economic value. I think that from a personal point of view, what you get from higher education depends a lot of what you put into the table, if you are a hard working intelligent guy and you have the ability to focus in what you are really good at, investing in higher education will give you a good ROI.

Why collage and not college? https://www.learnenglish.de/mistakes/collagevscollege.html

Why is American capitalized and asian isn't?

When you are designing an hypothesis test the term positive and negative are not so clear. For example you can test that mean weight of bags is greater than 5.0 kg or smaller than 5.0 kg, both test are different, and some times you can accept both greater and smaller than 5.0 kg. The philosophy of hypotesis test is not as clear as a standard tests for pregnancy. In other terms,in some cases the H0 hypothesis is symmetric (>= versus =<) and is not clear what a positive result should be, you have to state clearly what is the H0 hypothesis. In a pregnancy test everyone agree than H0 is that you are not pregnant, that is in my HO the semantic difference between Type 1 error and false negative.

The experiment you suggest seems promising. As a math teacher, I don't know about this field. But to suggest experiments that "could" prove or support and hypothesis is a good way to advance science. Thanks for our suggestions. As a Ph.D. math teacher I don't know about this field.

Don't know anything about RNA, but I wonder if some kind of small virus could be used to test the micro RNA transport hypothesis. Googling "micro RNA virus", first link is https://www.ncbi.nlm.nih.gov/pubmed/21431678 . The paper seems promising, since 2004, more than 200 microRNAs (miRNAs) have been discovered in double-stranded DNA viruses, mainly herpesviruses and polyomaviruses.This chapter aims to summarize our current knowledge of viral miRNAs, their targets and function, and the challenges lying ahead to decipher their role in viral biology.

Using microRNA virus to test the sperm transport mechanism seems to be a good idea. Glooging for "microRNA virus sperm transport" suggest the link: Intercellular Transport of MicroRNAs (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3580056/), This review brings into focus what is currently known and outstanding in a novel field of study with applicability to cardiovascular disease.

So, perhaps sperm transport of microRNA could be tested using some RNAvirus to produce cardiovascular desease.

This kind of attack relies on a low margin DNN, see (1), a low spectral norm of the input-output jacobian matrix guarantees good generalization error. So a one pixel attack exploits a weak eigenvalue (small absolute value) of the jacobian matrix.

So to create a one pixel attack, compute: 1)the eigenvalues of the jacobian of input-ouput matrix, 2) takes the the smaller eigenvalue lambda_1 3) compute or approximate the function lambda_1 = f(input) 4) compute j = argmax_{i=1..n} d(lambda_1)/d(input_i) at the point in which the spectral norm is maximum.

So to create the attack change the j-pixel in the points of the training set that has maximum (or high) jacobian matrix.

(1) https://arxiv.org/pdf/1605.08254.pdf

Given that DNN are deep, what a one pixel attack means is that one pixel change propagates through the map of features: one pixel => 0-level-feature change -> one 1-level feature change. So this attack relies in weak features that can easily propagate to next level of features. Hence, to defend against this attack the model should put a threshold on the ratio (sensitivity of features)/(number of pixels) and avoid features with high sensivity to easily propagate to the next level of the DNN. If features are not linearly related to input set, then correlation is not a measure of feature sensitivity and has nothing to say about the full DNN effect of such change in a pixel.

If in a DNN for label a cat we explore the group of movements of the animal cat (realistic movements available for a cat) we could relate the discriminative power of the DNN to the energy of the cat. The energy of the cat is related to the volume of the group of movements. A cat with zero energy has the identity group of movements (no movement), a hulk cat is able to alter many of her features, so a very power model is needed to identify a hulk cat. A person full of rage is able to change the color of her face, again energy alter training space. Sorry for using HN for thinking.

this paper https://openreview.net/forum?id=HJC2SzZCW suggest that sensivity is related to poor generalization power. To define derivative we need to use a natural parameter in such a way that it measures sensivity and also allow us to use methods from calculus and manifolds, such as parallel transport of features. How a DNN label a cat when is catching a rat.

We need to define the derivative of a deep model. I mean a way to measure how a model change when we change one pixel in the training data. Since pixel -> feature -> margin, we need to define the derivative with respect to a natural parameter, the natural parameter of the model has to defined ad hoc for every application. Perhaps the natural parameter encodes an uninformative prior. The intuition is to use information theory to see how the discriminative power of the model change when the training data is perturbed. So we need to measure the derivative of the added information. Fisher information seems to be related to this.