IPC presentation at ICAPS: your input

[Malte Helmert]

Dear IPC-2008 group,

the IPC results will be presented at ICAPS in two ways: in a talk, and
in the poster session. For both events, we would like to give at least
a brief overview of what the competing systems are, and for this we
need your input.

In the talk, we would like to have *one slide* per planner. In the
poster session, we will have one A0 poster for the planners, which
leaves us about one A5 sheet's worth of space per planner. For the
Americans, this is about half a letter-size page. (Note this is a
*poster*, so the font size will be large.) In both cases, we want to
mention the planner name and team members, plus a short high-level
description of the main features of the planner.

What we want from you is a short summary (about 4-6 bullet points) of
how the planner works. Don't include info on the setting (that you do
temporal planning, etc.) -- this will be clear from context. For
example, if your name is Jörg Hoffmann and your planner is called FF,
your description might look like this:

* does \emph{heuristic forward search} in the state space
* heuristic based on \emph{delete relaxations}, extends Bonet and
Geffner's additive heuristic
* novel local search algorithm: \emph{enforced hill-climbing}
* novel pruning technique to reduce branching factor: \emph{helpful
actions}

Please send me this information as plain text, and feel free to attach
any pictures that illustrate your points or are otherwise relevant for
the planner. (For example, for FF there's the highly relevant
http://members.deri.at/~joergh/ff/fieselschweif.gif.) High-res or
vector images are preferred for the poster. Keep in mind we don't have
much space, though.

If your planner participates in several tracks, please give *one
description per track*, since we will organize this by tracks. If you
have two closely related planners in the same track, you can give
separate descriptions or one per planner, whatever you prefer.

Best regards,

Malte