2020-06-26 16:15:20 (GMT)
Stream created by Simon Cruanes.
2020-06-26 16:16:6 (GMT)
Let's create one topic for each talk (that at least one person here attends and wants to share about)! :slight_smile:
2020-06-26 16:38:11 (GMT)
I will use this opportunity to promote my talks at PAAR(June 30. 13:30) and FSCD (July 2, 18:00)
2020-06-26 16:38:41 (GMT)
selfishly :slight_smile:
2020-06-26 16:50:8 (GMT)
Reminds me I have to my slides...
2020-06-26 18:48:35 (GMT)
Let me similarly promote our talks at PAAR (June 30th 15:00 and 16:00) and IJCAR (work of @Ahmed B ) (1st July 13:30 and 14:30, and 4th July 17:00)... it's going to be a busy week.
2020-06-26 20:12:36 (GMT)
Waldmann, Tourret, Robillard, and me (with Uwe talking) on 1 July at 14:00 at IJCAR, about a framework for proving dynamic completeness of calculi and given-clause-based provers.
conference version · technical report
(And in case you wonder about the reverse alphabetical authors' list: Uwe is bossed around by Sophie, who's bossed around by Simon, and guess who's the head honcho.)
2020-06-26 20:13:42 (GMT)
We should probably make one new topic per talk, but that can be done on the fly.
2020-06-29 9:40:0 (GMT)
I have created a whole steam just for PAAR where talks are grouped by sessions : #PAAR 2020 . The attendees of PAAR have been invited to register to Zulip to have discussions there. Lets see if it works, and feel free to use the "coffee breaks" topic to advertise your job offers and the like (@Giles ) today and tomorrow.
2020-06-29 9:44:27 (GMT)
Sophie, I did not seem to get PAAR Zoom coordinates
2020-06-29 9:44:47 (GMT)
Did you register for it?
2020-06-29 9:45:28 (GMT)
Anyway, I'll forward you the email. You can of course ignore the zulip invitation since you are already there!
2020-06-29 9:45:58 (GMT)
I thought FSCD/IJCAR registration is the same
2020-06-29 9:46:19 (GMT)
No, there is a separate registration link for all satellite events.
2020-06-29 9:46:21 (GMT)
but it maybe isn't> :innocent:
2020-06-29 9:46:36 (GMT)
Indeed ^^
2020-06-29 9:48:17 (GMT)
I'm happy that PAAR has such a stream. It's odd to not "see" other participants in anyway. There is the coffee break zoom too, but it was not heavily used just now.
2020-06-29 9:48:54 (GMT)
Well, PAAR only starts in a few hours...
2020-06-29 9:51:24 (GMT)
Right, but it wasn't heavily used by participants of other workshops either. ^^
2020-06-29 10:52:46 (GMT)
How do I find the coffee break room? And that reminds me - I need to make some coffee ;-)
2020-06-29 10:58:9 (GMT)
The Zoom room is the last Zoom link in the "[FSCD-IJCAR Workshops 2020] Day 1 - Practical information" mail from yesterday.
2020-06-29 11:14:8 (GMT)
When I click on that link it leads me here.
Can you see the PAAR room?
2020-06-29 11:14:37 (GMT)
You can subscribe to it by clicking "Add streams". I had to do that.
2020-06-29 11:15:35 (GMT)
ok thanks :slight_smile:
2020-06-29 11:40:9 (GMT)
Aha! I did not get that Email - I registered for IJCAR ages ago, but for PAAR only tonight, after I found out that it needs a separate step....
2020-06-29 11:53:21 (GMT)
Same here, no email. Could somebody post the content of the email here?
2020-06-29 11:54:25 (GMT)
Hi Petar Vukmirovic,
Thank you for registering for "Workshop on Practical Aspects of Automated Reasoning".
Please submit any questions to: fscd-ijcar-4@univ-paris13.fr
Date Time: Jun 29, 2020 12:30 PM Paris
Every day, until Jun 30, 2020, 2 occurrence(s)
Jun 29, 2020 12:30 PM
Jun 30, 2020 12:30 PM
Please download and import the following iCalendar (.ics) files to your calendar system.
Daily: https://zoom.us/meeting/attendee/tJwkd-Gtpz8sGNcTx4G_rTVMYSpGBgP_9KfF/ics?user_id=oWEHZEb7QIaA4KzvD7M5AA
Join from PC, Mac, Linux, iOS or Android: Click Here to Join
Note: This link should not be shared with others; it is unique to you.
Add to Calendar Add to Google Calendar Add to Yahoo Calendar
Or iPhone one-tap
US: +13017158592,,98908594623# or +13126266799,,98908594623#
Or Telephone:
Dial(for higher quality, dial a number based on your current location):
US: +1 301 715 8592 or +1 312 626 6799 or +1 929 436 2866 or +1 253 215 8782 or +1 346 248 7799 or +1 669 900 6833
Meeting ID: 989 0859 4623
International numbers available: https://zoom.us/u/aeehjuJETf
Or an H.323/SIP room system:
H.323:
162.255.37.11 (US West)
162.255.36.11 (US East)
115.114.131.7 (India Mumbai)
115.114.115.7 (India Hyderabad)
213.19.144.110 (EMEA)
103.122.166.55 (Australia)
209.9.211.110 (Hong Kong SAR)
64.211.144.160 (Brazil)
69.174.57.160 (Canada)
207.226.132.110 (Japan)
Meeting ID: 989 0859 4623
SIP: 98908594623@zoomcrc.com
You can cancel your registration at any time.
2020-06-29 11:54:54 (GMT)
[FSCD-IJCAR Workshops 2020] Day 1 - Practical information
2020-06-30 11:49:47 (GMT)
Could someone post the Day 2 info? I must have registered too late and am not getting any emails.
2020-06-30 11:50:29 (GMT)
It's the same zoom links I think.
2020-06-30 11:51:27 (GMT)
Dear participants,
2020-06-30 17:29:44 (GMT)
For those who don't receive emails:
[FSCD-IJCAR 2020] Practical information * ZOOM INVITATION INSIDE *
2020-06-30 18:22:0 (GMT)
IJCAR starts with a vengeance tomorrow. First section is very interesing :)
2020-07-01 11:48:5 (GMT)
On track B, Lochbihler & Traytel are putting on quite a show. It's worth watching, if not for the content, for the presentation techniques.
2020-07-01 11:54:1 (GMT)
I had not realized there were parallel sessions :( I'll check that talk later then :)
2020-07-01 11:55:41 (GMT)
Where are the recordings available?
2020-07-01 11:55:56 (GMT)
2020-07-01 11:56:11 (GMT)
They are not there yet but that is where they should be.
2020-07-01 11:56:23 (GMT)
Or do you want to talk about the live streams?
2020-07-01 11:57:1 (GMT)
Where did you find the youtube link?
2020-07-01 11:57:4 (GMT)
I was looking for it
2020-07-01 11:57:24 (GMT)
I searched for fscd-ijcar 2020 in youtube
2020-07-01 11:57:29 (GMT)
Or do you want to talk about the live streams?
No, the youtube link is what I was looking for, too.
2020-07-01 11:57:41 (GMT)
But there were going to be live ones?
2020-07-01 11:57:50 (GMT)
in sync with the zoom sessions
2020-07-01 11:58:21 (GMT)
Here is the stream link for session IJCAR A: https://youtu.be/lwDMi0SfVzg
2020-07-01 12:27:43 (GMT)
On the fscd-ijcar 2020 website, I found an even better link for the live streams (you can access any of the m from here): https://www.youtube.com/playlist?list=PLl1dj5prwUJzc1P5_k4GvlUrmwo3OWrUU
2020-07-01 12:42:52 (GMT)
Ugh, I thought that one was automatically muted back after asking a question.
2020-07-01 12:43:4 (GMT)
This zoom setup used for IJCAR I had never used before
2020-07-01 12:43:35 (GMT)
Do you mean the webinars?
2020-07-01 12:43:49 (GMT)
Yeah, if that's what's called
2020-07-01 12:44:6 (GMT)
there is some construction going on close to here and it was noisy
2020-07-01 12:44:9 (GMT)
I think it puts a lot of charge on the shoulders of the host
2020-07-01 12:44:23 (GMT)
Yeah, it seems so. Jeremy seemed to be struggling a bit
2020-07-01 13:1:1 (GMT)
the coffee break track does not seem very popular
2020-07-01 13:5:9 (GMT)
I kind of saw this coming and cowardly ignored all requests to become a session chair. ;)
2020-07-02 17:14:2 (GMT)
CADE is gonna be in Pittusburg, had no idea
2020-07-02 17:14:27 (GMT)
A good opportunity to talk to Marijn!
2020-07-03 12:2:53 (GMT)
So @Pascal Fontaine 's politeness paper got best paper award. Congrats!
By the way, the chat message said co-winner. Who else got it?
2020-07-03 12:4:40 (GMT)
Joshua Brakensiek, Marijn Heule, John Mackey and David Narvaez for solving Keller's conjecture
2020-07-03 12:5:38 (GMT)
Thanks!
2020-07-03 13:53:31 (GMT)
The solution was like always in life, be lazy
Ruzica Piskac
:smiley:
2020-07-03 13:55:45 (GMT)
I know this is kind of a classic in our field, but it is nice to be reminded of it in a great talk!
2020-07-03 13:58:45 (GMT)
Don't do today what you can do tomorrow. Pretty much how I lead my professional life
2020-07-03 13:59:57 (GMT)
I'd noticed. ;)
2020-07-03 14:0:34 (GMT)
:P
2020-07-03 14:0:59 (GMT)
You seemed to be shocked when I proposed we start writing our CADE paper two months before the deadline.
2020-07-03 14:3:13 (GMT)
hahaha it's good practice though! I should do that more
2020-07-03 14:3:34 (GMT)
You mean after the deadline, right?
2020-07-03 14:11:18 (GMT)
I'm working on my next CADE submissions right now. Is that after the deadline or before the deadline?
2020-07-03 14:14:23 (GMT)
I hope it stars with "nun" and ends with "ku" :joy: (just kidding)
2020-07-03 14:16:17 (GMT)
It starts with "su" and ends with "ic". ;)
2020-07-03 14:18:4 (GMT)
I think we perform too many needless ArgCong inferences and want to curb that.
2020-07-03 16:39:23 (GMT)
Sophie Tourret said:
So Pascal Fontaine 's politeness paper got best paper award. Congrats!
By the way, the chat message said co-winner. Who else got it?
Congratulations :slight_smile:
2020-07-04 12:55:20 (GMT)
Rereading your question in the Q&A @Sophie Tourret I think I misunderstood it a bit. I thought you were asking in creating a grammar for a specific problem but you were wondering for new logics, presumably for the ones we do not have a default grammar for?
2020-07-04 12:55:40 (GMT)
yes
2020-07-04 12:56:2 (GMT)
The answer does not change much though, it's easy. One just has to determine what are the operators of the theory and build the proper datatype (we use datatypes to encode the grammars internally)
2020-07-04 12:56:11 (GMT)
By the way, thanks for the nice talk! :slight_smile:
2020-07-04 12:56:17 (GMT)
The gap is basically because we do it on a on-demand basis
2020-07-04 12:56:27 (GMT)
Thanks, I'm glad you liked it :)
2020-07-04 12:57:6 (GMT)
It is very interesting how you use counter-examples to filter the candidates.
2020-07-04 12:58:23 (GMT)
Yes, it's what really makes a difference (the rest of the process is standard SyGuS). This is tightly tied with other work in SyGuS solving for using counterexamples to better guide the search. But while these approaches are general-purpose we did this one to leverage the specific shape of the abduction problem, the fact that you have two somewhat independent things to meet (consistency with the axioms and entailing the goal)
2020-07-04 14:24:1 (GMT)
Hi Haniel, great talk and nice work!
Have you got experience with allowing disjunction in the solutions? Is that allowed and what about quantification?
2020-07-04 14:33:45 (GMT)
Thanks, @Renate Schmidt !
2020-07-04 14:33:48 (GMT)
We can allow disjunctions in the solutions, yes. In this particular work we generate disjunctive solutions when generating solutions incrementally and optimizing for logical weakness. But when finding for a single abduct we could include disjunctions directly (it's just a matter of including it or not in the grammar).
2020-07-04 14:35:5 (GMT)
We can also include quantifiers. The difficulty there is that the search space increases significantly and that the satisfiability check for verifying candidate solutions (when the previous counterexamples don't rule them out) can be undecidable. But it is supported
2020-07-04 14:35:47 (GMT)
(in the paper we have a note specifically about that)
2020-07-04 14:36:40 (GMT)
Cool, thanks!
2020-07-04 14:36:56 (GMT)
An interesting note also is that when generating abducts in QF_UFLIA the background theory for the verification of candidate solutions was higher-order, but it works since CVC4 supports it :)
2020-07-04 14:37:21 (GMT)
:+1: