AI Astronomer Aids Effort To Analyze Galaxies

kkleiner writes "Scientists are teaching an artificial intelligence how to classify galaxies imaged by telescopes like the Hubble. Manda Banerji at the University of Cambridge, along with researchers at University College London, Johns Hopkins, and elsewhere, has succeeded in getting the program to agree with human analysis at an impressive rate of more than 90%. Banerji used data from Galaxy Zoo, a massive online project that has used more than 250,000 volunteers to analyze more than 60 million galaxies. The new automated astronomer will help with even larger analytical projects on the horizon, taking care of trivial classifications and leaving the tough cases to humans."

Comments

better than humans?

[samwise098]

I wonder how this program compares to a human doing the same job If given the same "training" I wonder how many humans would get a 90% agreement rate looking at the same data.

Name

[kellyb9]

I would think with a name like Al Astronmer your career choices would be limiting. I guess I was right.

Original paper

[JoshuaZ]

The paper discussing this work is http://arxiv.org/abs/0908.2033. They appear to be using a pretty standard neural network approach (disclaimer: I don't have much background in neural nets at all. I'm just going off of how they were described in the last class I took that discussed them.) This is part of a very general pattern where programs have done a lot of work that we would think could only be done by people. Other examples include the computerized proof of the Robbins conjecturehttp://en.wikipedia.org/wiki/Robbins_conjecture. TFA lists a few examples as well which are in more applied areas.

Not a troll - actually a very good point

[Anonymous Coward]

Let's say the system of scientific paradigms and their rise and fall is about finding outliers and surprising results that cannot easily be explained by current models.

If this is the case, then using a statistical system to classify observations has the danger that these "outliers" simply get classified in existing categories and whatever abnormality they represent thereby ignored.

An "AI researcher" must therefore have very explicit programming to set anything with even the slightest degree of abnormality aside for human evaluation. If it's set to classify anything and everything according to preset rules, it's actually mostly destructive to good science.

I have seen the future

[Locke2005]

If this technologies works for classifying galaxies, perhaps next we could put it to work classifying porn on the web!

pros and cons of this approach

[Anonymous Coward]

PRO:

Using neural networks allows for graceful degradation when classifying galaxies by indicating to what degree it believes this galaxy is similar to other galaxies of this type (that it has been trained on). A threshold can be set so that if confidence falls below this threshold, the image is flagged for human intervention.

CONS:

Neural nets are largely black boxes. They use learned statistical relationships to classify images, but they're unable to provide an explanation as to why they made the decision that they did.

Re:What's the point?

[Daniel Dvorkin]

There are a couple of answers to your question. The first is the answer to the more general question, "Why study the universe at all?" and the answer is "Because it's there." We want to understand the processes by which the universe we see around us was formed, what it's like now (to the degree that "now" has any meaning on cosmological scales) and where it's going. It is an awe-inspiring place, and becomes more so the more we learn about it.

The second, with respect to the study of the Milky Way, is that we learn a lot about our galaxy by studying other galaxies. We don't have a good vantage point for studying the Milky Way, for obvious reasons. Hell, it wasn't until quite recently that we even knew what shape it was (barred spiral vs. plain spiral.) With the enormous number of galaxies out there, many of them similar to our own, at a variety of viewing angles from Earth, we can get a much better idea of what's going on in our own neighborhood than we could by restricting our observations to the Milky Way alone.

Re:What's the point?

[$RANDOMLUSER]

Because the farther away they are, the farther back in time we're looking. By collecting images of galaxies at different stages of evolution (and different types of collisions) cosmologists are able to form a much better picture of how galaxies (and the universe in general) form and evolve.