Archive for inference

Six New Publications at the Open Journal of Astrophysics

Posted in OJAp Papers, Open Access, The Universe and Stuff with tags , , , , , , , , , , , , , , , , , , , , , , , , , on October 12, 2024 by telescoper

Regular readers of this blog (both of them) will have noticed that I didn’t post an update of activity at the Open Journal of Astrophysics last weekend. Despite having accepted several papers for publication in the preceding week, no final versions had made it onto the arXiv. We can’t published a paper until the authors post the final version, so that meant a bit of a backlog developed. This week included one day with no arXiv update (owing to a US holiday on Tuesday 8th October) and a major glitch on Crossref on Thursday which delayed a couple, but even so we’ve published six papers which is the most we’ve ever managed in a week. This week saw the publication of our 200th article; the total as of today is 202.  The count in Volume 7 (2024) is now up to 87; we have four papers in the queue for publication so we should pass 90 next week if all goes well.

In chronological order, the six papers published this week, with their overlays, are as follows. You can click on the images of the overlays to make them larger should you wish to do so.

First one up, published on Monday 7th October 2024 is “z~2 dual AGN host galaxies are disky: stellar kinematics in the ASTRID Simulation” by Ekaterina Dadiani (CMU; Carnegie Mellon U.) Tiziana di Matteo (CMU), Nianyi Chen (CMU), Patrick Lachance (CMU), Yue Shen (U. Illinois at Urbana-Champaign), Yu-Ching Chen (Johns Hopkins U.), Rupert Croft (CMU), Yueying Ni (CfA Harvard) and Simeon Bird (U. California Riverside) – all based in the USA. The paper, which is in the folder marked Astrophysics of Galaxies describes a numerical study of the morphology of AGN host galaxies containing close pairs of black holes.

Here is a screen grab of the overlay, which includes the abstract:

 

You can find the officially accepted version of the paper on the arXiv here.

The second paper to announce, published on 8th October 2024, is “Origin of LAMOST J1010+2358 Revisited” by S.K. Jeena and Projjwal Banerjee of the Indian Institute of Technology Palakkad, Kerala, India. This paper discusses  the possible formation mechanisms for Very Metal Poor (VMP) stars and the implications for the origin of LAMOST J1010+2358 and is in the folder marked Solar and Stellar Astrophysics.

You can see the overlay here:

The accepted version of this paper can be found on the arXiv here.

The third paper is very different in both style and content: “Assessing your Observatory’s Impact: Best Practices in Establishing and Maintaining Observatory Bibliographies” by Raffaele D’Abrusco (Harvard CfA and 14 others; the Observatory Bibliographers Collaboration) and is in the folder marked Instrumentation and Methods for Astrophysics. It presents discussion of the methods used by astronomical observatories to construct and analyze bibliographic databases. The overlay is here:

(This one gave me a rare opportunity to use the library of stock images that comes with the Scholastica platform!) The officially accepted version can be found on arXiv here.

The fourth paper, also published on 8th October 2024, and our 200th publication, is in the folder marked Cosmology and NonGalactic Astrophysics, and is called “CombineHarvesterFlow: Joint Probe Analysis Made Easy with Normalizing Flows“. The authors are Peter L. Taylor, Andrei Cuceu, Chun-Hao To, and Erik A. Zaborowski of Ohio State University, USA. The article presents a new method that speeds up the sampling of joint posterior distributions in the context of inference using combinations of data sets. The overlay is here

You can find the officially accepted version of this paper here.

The fifth paper in this batch is “Estimating Exoplanet Mass using Machine Learning on Incomplete Datasets” by Florian Lalande (Okinawa Institute of Science and Technology), Elizabeth Tasker (Institute of Space and Astronautical Science, Kanagawa) and Kenji Doya (Okinawa); all based in Japan. This one was published on 10th October 2024 in the folder marked Earth and Planetary Astrophysics. It compares different methods for inferring exoplanet masses in catalogues with missing data

 

 

You can find the official accepted version on the arXiv here.

Finally for this week we have “Forecasting the accuracy of velocity-field reconstruction” by Chris Blake and Ryan Turner of Swinburne University of Technology, Melbourne, Australia. This was also published on 10th October 2024 and is in the folder marked Cosmology and NonGalactic Astrophysics. The paper describes a numerical study of the reliability and precision of different methods of velocity-density reconstruction. The overlay is here

You can find the officially-accepted version on arXiv here.

That’s it for now. We have published six papers, with a very wide geographical spread of authors, and in five of the six astro-ph categories we cover. I think it’s been a good week!

1 Comment »

Machine Learning in the Physical Sciences

Posted in The Universe and Stuff with tags , , , , , on March 29, 2019 by telescoper

If, like me, you feel a bit left behind by goings-on in the field of Machine Learning and how it impacts on physics then there’s now a very comprehensive review by Carleo et al on the arXiv.

Here is a picture from the paper, which I have included so that this post has a picture in it:

The abstract reads:

Machine learning encompasses a broad range of algorithms and modeling tools used for a vast array of data processing tasks, which has entered most scientific disciplines in recent years. We review in a selective way the recent research on the interface between machine learning and physical sciences.This includes conceptual developments in machine learning (ML) motivated by physical insights, applications of machine learning techniques to several domains in physics, and cross-fertilization between the two fields. After giving basic notion of machine learning methods and principles, we describe examples of how statistical physics is used to understand methods in ML. We then move to describe applications of ML methods in particle physics and cosmology, quantum many body physics, quantum computing, and chemical and material physics. We also highlight research and development into novel computing architectures aimed at accelerating ML. In each of the sections we describe recent successes as well as domain-specific methodology and challenges.

The next step after Machine Learning will of course be Machine Teaching…

Leave a comment »

Beyond Falsifiability: Normal Science in a Multiverse

Posted in The Universe and Stuff with tags , , , , , , on January 17, 2018 by telescoper

There’s a new paper on the arXiv by Sean Carroll called Beyond Falsifiability: Normal Science in a Multiverse. The abstract is:

Cosmological models that invoke a multiverse – a collection of unobservable regions of space where conditions are very different from the region around us – are controversial, on the grounds that unobservable phenomena shouldn’t play a crucial role in legitimate scientific theories. I argue that the way we evaluate multiverse models is precisely the same as the way we evaluate any other models, on the basis of abduction, Bayesian inference, and empirical success. There is no scientifically respectable way to do cosmology without taking into account different possibilities for what the universe might be like outside our horizon. Multiverse theories are utterly conventionally scientific, even if evaluating them can be difficult in practice.

I’ve added a link to `abduction’ lest you think it has something to do with aliens!

I haven’t had time to read all of it yet, but thought I’d share it here because it concerns a topic that surfaces on this blog from time to time. I’m not a fan the multiverse because (in my opinion) most of the arguments trotted out in its favour are based on very muddled thinking. On the other hand, I’ve never taken seriously any of the numerous critiques of the multiverse idea based on the Popperian criterion of falsifiability because (again, in my opinion) that falsifiability has very little to do with the way science operates.

Anyway, Sean’s papers are always interesting to read so do have a look if this topic interests you. And feel free to comment through the box below.

4 Comments »