Dear CASPERites, SETI-interested people,

Happy New Year! 

The purpose of this email is to advertise a short questionnaire 
(https://goo.gl/forms/yH3LlTrIqtCZIEXN2 
<https://goo.gl/forms/yH3LlTrIqtCZIEXN2>) to help us (George Hobbs, me, ++) set 
up a meeting relating to “finding the unknown” in astronomical data sets.

I think that we’d all agree that astronomy data sets are becoming massive and 
processing those data sets is becoming more and more of a challenge.  In 
response to this, we develop pipelines for finding specific sources of interest 
- for example, simple objects in images, pulsars/FRBs in radio time-series data 
sets, and known classes of transients/events in optical time-domain data.  
However, as our algorithms become more and more optimised for particular 
sources, we will lose our ability to discover anything currently unknown and 
unexpected.  This is a big issue for future telescopes such as the SKA, LSST 
etc, which will produce such large data volumes that much of the data 
processing will need to be done in near real time, and the raw data deleted.

Fast radio bursts (FRBs) were discovered in a reprocessing of archival Parkes 
data using an algorithm that enabled more parameter space to be searched. 
Pulsars themselves were, of course, discovered entirely serendipitously by 
Jocelyn Bell looking by eye at chart recorder output.  How can such discoveries 
be made with future telescopes?

We (George Hobbs, me, ++) would like to organise a meeting/workshop/conference 
on this topic: how to discover the unknown.  George has already touched base 
with a few people and has some ideas, but it isn’t clear whether we should 
organise a short workshop on a well-defined topic (e.g., “finding bright, 
transient events, in single-dish, radio astronomy, time-series data”), or have 
a slightly more general meeting (e.g., “finding the unknown in radio astronomy 
data sets”), or a very general one (“how to find anomalies in a given data set” 
- which may not even necessarily have the word “astronomy” in it!).

It has been suggested that an SOC be set up, but we think it is important to 
define the scope of the meeting (or meetings!) first before selecting the SOC.

We have set up a short questionnaire available from 
https://goo.gl/forms/yH3LlTrIqtCZIEXN2 
<https://goo.gl/forms/yH3LlTrIqtCZIEXN2>.  If you are interested in this, would 
you be willing to fill in the questionnaire as soon as convenient (we’ll close 
the questionnaire on 14th Jan)? This will greatly help with planning what 
should be a fantastic event.   We’ll use the results from the questionnaire to 
(1) better scope the meeting and (2) to identify a set of people to be on the 
SOC/LOC.   Note that the results of the questionnaire will be shared with the 
final SOC members.

Please forward this message to anyone else you know who may be interested.  
We're planning on getting input from astronomers, computer experts, hardware 
engineers, algorithm experts, cloud/HPC teams and industry, and people from the 
LIGO, medical imaging, public outsourcing, etc.

(apologies if you receive this email more than once!)

Best,
Vikram

-- 
You received this message because you are subscribed to the Google Groups 
"[email protected]" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].

Reply via email to