You can subscribe to this list here.
2010 
_{Jan}

_{Feb}

_{Mar}

_{Apr}

_{May}

_{Jun}

_{Jul}

_{Aug}

_{Sep}
(1) 
_{Oct}
(8) 
_{Nov}

_{Dec}


2011 
_{Jan}

_{Feb}

_{Mar}
(5) 
_{Apr}
(4) 
_{May}
(2) 
_{Jun}

_{Jul}
(7) 
_{Aug}
(34) 
_{Sep}
(11) 
_{Oct}
(25) 
_{Nov}
(9) 
_{Dec}
(19) 
2012 
_{Jan}
(3) 
_{Feb}

_{Mar}
(7) 
_{Apr}
(13) 
_{May}
(13) 
_{Jun}
(24) 
_{Jul}
(16) 
_{Aug}
(2) 
_{Sep}

_{Oct}
(14) 
_{Nov}
(34) 
_{Dec}
(9) 
2013 
_{Jan}
(38) 
_{Feb}
(18) 
_{Mar}
(17) 
_{Apr}
(17) 
_{May}
(36) 
_{Jun}
(3) 
_{Jul}
(4) 
_{Aug}
(25) 
_{Sep}
(25) 
_{Oct}
(15) 
_{Nov}
(29) 
_{Dec}
(9) 
2014 
_{Jan}
(40) 
_{Feb}
(84) 
_{Mar}
(25) 
_{Apr}
(47) 
_{May}
(28) 
_{Jun}
(21) 
_{Jul}
(19) 
_{Aug}
(27) 
_{Sep}
(16) 
_{Oct}
(68) 
_{Nov}
(33) 
_{Dec}
(15) 
2015 
_{Jan}
(15) 
_{Feb}
(16) 
_{Mar}
(2) 
_{Apr}
(11) 
_{May}
(16) 
_{Jun}
(65) 
_{Jul}
(26) 
_{Aug}
(22) 
_{Sep}
(18) 
_{Oct}
(40) 
_{Nov}
(12) 
_{Dec}

S  M  T  W  T  F  S 





1
(2) 
2
(2) 
3

4

5

6

7
(2) 
8

9

10

11

12

13

14

15

16
(1) 
17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

From: Michal Certicky <certicky@fm...>  20120316 01:32:25

Hello Zahra, There is a specific kind of inductive learning called "action learning", that I have recently implemented using oClingo. You may want to take a look at the following paper: M. Čertický. Action Learning with Reactive Answer Set Programming: Preliminary Report. In Proceedings of The Eighth International Conference on Autonomic and Autonomous Systems (ICAS 2012), 2012. It can be downloaded from http://breakk.noip.org/files/publications/icas2012.pdf and should contain both the explanation, and a link to actual implementation. I hope it helps. Michal Certicky On 16 December 2011 16:38, Martin <mjb@...> wrote: > > On Fri, 20111216 at 07:25 0800, Zahra Vaseqi wrote: > > > > Hi everyone, > > > > > > I am trying to figure out how can I implement "inductive learning", > > which is mentioned among potential applications of the approach, using > > oClingo. > > > > I would appreciate it if you could give me some hints on that. A small > > example would be great. > > I know Alessandra Russo at Imperial did some work on building an > inductive logic programming system on top of an answer set solver; > perhaps this paper might be a starting point: > > http://ilp11.doc.ic.ac.uk/short_papers/ilp2011_submission_20.pdf > > HTH > > Cheers, >  Martin > > > >  > Learn Windows Azure Live! Tuesday, Dec 13, 2011 > Microsoft is holding a special Learn Windows Azure training event for > developers. It will provide a great way to learn Windows Azure and what it > provides. You can attend the event by watching it streamed LIVE online. > Learn more at http://p.sf.net/sfu/mswindowsazure > _______________________________________________ > Potasscousers mailing list > Potasscousers@... > https://lists.sourceforge.net/lists/listinfo/potasscousers 
From: Roland Kaminski <kaminski@cs...>  20120307 19:38:49

From: Andrea Peano <andrea.peano@un...>  20120307 18:37:02

Dear all, I've an optimization issue about the maximization of a min value. This is my first step in ASP and maybe it is too long! I have N different context and for each context I have a value which is calculated through a #sum aggregator: value(V, Context) : V = #sum [ something(C, Context) = C]. So, I have to maximize the minimum value over all contexts: min_value(MIN) : MIN = #min [ value(V, Context) = V]. #maximize[min_value(MIN)=MIN@...]. In this case (in my case) the grounder generates a ground atom like: min_value(#supremum) : ... Thus, Gringo returns this error: ERROR: cannot convert #supremum to integer in: maxmin.lp:59:13: #maximize[min_dem(Dmin)=Dmin@...] Is mine a wrong way to model an optimization problem like this?? Any suggestions?? Thank you, Andrea 
From: Roland Kaminski <kaminski@cs...>  20120302 10:22:03

From: Adam Smith <adam@ad...>  20120302 06:56:32

On Thu, Mar 1, 2012 at 1:29 PM, Torsten Schaub <torsten@...> wrote: > what about taking a solution, turning it into integrity constraints, and > relaunching the solver with the extended problem and relaxed #hide/#show > statements? That was my first guess, but I got scared off by the potential cost of relaunching with these constraints multiplied by the number of expected interesting solutions (hundreds to thousands, but not trillions). 
From: Torsten Schaub <torsten@cs...>  20120301 21:29:13

Adam, what about taking a solution, turning it into integrity constraints, and relaunching the solver with the extended problem and relaxed #hide/#show statements? Cheers torsten 
From: Adam Smith <adam@ad...>  20120301 07:47:50

Hi all, I just put together a messy hack, and I'm wondering if any of you have run into a similar situation. I've got solver for a complex puzzle game, and I'm interested in enumerating all of the solutions that have distinct projection into an abstracted representation. Of all of the crazy ways to place pieces down on a board to form a solution, I only care about the number of distinct relationships pieces can have (imagine a lineofsight graph). Putting it in Chess terms, if you have a king, queen and pawn, and I have a king, then there are only three highlevel solutions for putting me in check (queen attacks, pawn attacks, both pawn and queen attack simultaneously), despite there being many many many ways of arranging the pieces on the game's 8x8 grid that result in check. Clingo has a nice "project" flag for enumerating only those solutions that are have a unique projection onto the terms that are "#show"n. This does exactly what I want with respect to limiting the space of solutions. However, it hides the concrete piece positions from me  and I want these to draw a nice picture! Here is how I printed out a hidden predicate via Lua despite it being hidden under projection: %% #begin_lua function onModel() Assignment.begin("p",1) while Assignment.next() do print("p",Assignment.args()[1],Assignment.isTrue()) end end#end_lua.{p(1..8)}. zz : p(1). #hide. #show zz. %% If you ask Clingo for all solutions, you get all 256 ways of flipping the bits of p/1. If you add "project", you only see two solutions: with and without zz/0. The Lua hack gives me access to the rest of the certificate, but it feels dirty. How would you solve this in a pretty way? Related, I wonder if this is showing the need for a "#project" directive that exposes the projection feature in a slightly more transparent and "#show"like manner. Adam PS. Wouldn't it be amazing if there were a "#universal" directive for 2QBF modeling? 