xtacy

It's interesting you mention similarities with machine learning, when in fact, chronologically speaking, the formalisation of Quantum Mechanics used (complex) tools from Hilbert Space.

Also highly recommend Terry Tao's talk on the cosmic distance ladder:

https://terrytao.wordpress.com/2010/10/10/the-cosmic-distanc...

As the above post says, the main known motive is to diversify Saudi's sovereign wealth away from Oil as their primary investment.

His push into social sciences did not seem very childish to me; it was more like a plea to have more explicit models. I haven't come across his editorials for econometricians. Could you post some references that I can read?

I have, and I really like it. If you read some of the historical debates between Judea Pearl and Don Rubin (of the Rubin Causal Model fame), and also Andrew Gelman (on his blog: http://andrewgelman.com/category/causal-inference/) you will have a much better appreciation of the nuances that Pearl was trying to push forth.

What I like about SEM and graphical model formulation is that it makes the model explicit and easy to communicate. It compactly encodes many hypotheses that you can test on your data.

I am not a native English speaker, but quite familiar with the use of "at (large) scale" in similar context.

It means that one has deployed it/is using the tool with a non-trivial (could be anywhere from few 100s to few 1000s of machines) amount of CPU and/or data. In the context of serving, "large scale" could also mean the number of queries/second hitting the serving layer.

Does https://www.elph.com work for anyone?

While that's technically correct, rent is often at least one order of magnitude more than the cost for food and clothes.

Expanding on the comment below, it's an unfortunate combination of many prevailing issues.

1. One of the things Prop 13 does is to cap the maximum rate at which a property tax grows per annum to 2%. However, due to housing demand, the property values grow at a much faster rate (7% is not unheard of). The property tax can change when there is a sale.

2. Bay area prices have skyrocketed.

3. The peninsula topography and zoning laws make it hard to expand housing capacity.

So, if you are an aging home owner, sure you can cash out of a sale. But where would you go? Unless you can afford another place in the same area, existing home owners do not have incentives to sell their property, which results in a very low inventory. And, low inventory further puts pressure on the prices and drives it up.

Off-topic question: Does anyone know which software the authors used to create the illustrations?

It's still low by historical standards: https://fred.stlouisfed.org/series/FEDFUNDS

From the project-zero bug (https://bugs.chromium.org/p/project-zero/issues/detail?id=13...), it looks like the first discussion on the issue dates back to July 3, with a working exploit posted just yesterday.

I believe it's covered: Topic 6.

The paper is probably OCRed.

It's unfortunate that such a scheme exists, which pollutes the youth of the nation. I too received several calls from an "unknown" number trying to extort money by having me install some anti-virus software to fix a non-existent computer issue.

This week's Kurzgesagt episode covered CRISPR in an easy to understand fashion: https://www.youtube.com/watch?v=jAhjPd4uNFY.

I tried browsing the Darpa challenge's website to know more, but I couldn't find any information. Could someone please post a link to a detailed description of the challenge?

Not quite, but as the paper says:

    Corollary:  No  probabilistic  regular  grammar
    exhibits criticality.

    In the next section, we will show that this statement is
    not true for context-free grammars (CFGs).

That is, there exists CFGs that exhibit criticality. Programming languages are often parsed by CFGs, so it's likely that some programming languages exhibit the same criticality structure as natural languages.

I've switched to zsh and it works quite well with multiple sessions.

I partly agree with you, but I wouldn't go to the extent of using "easily" in this phrase --

> Statistical dependence can be determined easily if you know the distributions,

Even if you know the distribution, a statistical test will make Type-I/II errors that you would have to take care of.

Actually, I find the text linked above hard to understand, without properly defining 'c.' What's the sample space?

In general, my sentiments are with the xkcd comic strip, but nothing more. Pearl's theories lay a firm foundation for communicating a causal hypothesis and manipulating it algebraically, but the true tests of causal hypothesis are:

- Experimental evidence

- The predictions it makes, in cases where experiments are hard to perform (e.g., in physics, when we make certain causal conjectures about how the universe works).

> But even they do not admit Pearson (or any other type) of correlation, but rather the nebulous "statistical dependence".

The notion of "Statistical dependence" is not nebulous. X and Y are independent if the joint distribution factorises as

    p(X, Y) = p(X) p(Y)

> even they do not admit Pearson (or any other type) of correlation,

Precisely. They operate purely in probabilistic dependence/independence terminology, from a theoretical point of view.

Not quite -- in your reference frame, you cross the blackhole's event horizon and go inside, but an observer (who is outside the event horizon) will only see you reach the event horizon and slowly "fade away."

Kudos on the polished project release. I am relatively new to go, and I am curious about the technology stack behind such a webapp. How does it work under the hood?

- How do you develop such web apps with html, css, javascript, go, etc. all interacting with each other?

- How are static assets packaged in a single binary?

- Any simple tutorial or stack walkthrough you would recommend me reading?

thanks!

I think what the article meant to say was that, for the same number of samples, the p-value when you have delta=50 (the difference between groups), stdev=10 is the same as the p-value of delta=0.5, stdev=0.1.

Depending on the study, finding a delta whose p-value is significant does not necessarily mean that the size of the effect (i.e., delta) might be significant enough to be useful.

You are also normalising the difference by the variance, so the t-statistic has no units.

Very nice visualisation!

It would be good to gracefully handle large graphs. I typed in "torvalds/linux" and my tab froze!

Nice initiative. Cheap serde and cross-language compatibility with an eye towards data scan intensive workloads is an important component!

Have you folks considered Supersonic engine from Google, which was designed with similar (but not as extensive as Arrow) goals in mind?

https://github.com/google/supersonic

I too found Pearl's book hard to navigate on first attempt. Do not let that stop you! After a hiatus, I stumbled upon this blog post [1], which explained the core ideas in Pearl's framework beautifully in a simple language. My advice is to persist, fill any holes in fundamentals (mostly basic probability), and persist. After working out the examples in the blog post on paper and contrasting it to other ideas out there (potential outcome framework), it became quite clear what Pearl was trying to articulate.

Pearl is also an enthusiastic speaker. You can search for his talks online at various venues (Stanford, Microsoft Research, etc.) to learn more.

[1] http://www.michaelnielsen.org/ddi/if-correlation-doesnt-impl...

Thanks for references.

I've read the topology and data paper; it lays down motivations for TDA, but it doesn't quite connect it to existing literature on dimensionality reduction and manifold learning and explain -- "Here's something you can learn by using tools from TDA, but not existing methods." The best I could see was that it produces results similar to existing methods.

I find the topic intriguing, but can someone who is well versed with both topology and machine learning comment on what is the key innovation here?

On first glance, the methods here seem a lot like the toolbox of dimensionality reduction techniques (PCA, spectral embedding, or more general manifold learning, etc.) from machine learning literature.

What specific insights from the field of topology have helped further our understanding of data that we missed earlier?