From predragp at andrew.cmu.edu  Wed Feb  5 17:14:09 2014
From: predragp at andrew.cmu.edu (predragp at andrew.cmu.edu)
Date: Wed, 5 Feb 2014 17:14:09 -0500
Subject: [Research] Block Partitions of Sequences
Message-ID: <1ec851db73bb2a10bcc27b0ca82e9350.squirrel@webmail.andrew.cmu.edu>

Dear All,

I would just like to bring to your attention that tomorrow there would
be an interesting talk at Algorithms, Combinatorics and Optimization
Seminar of mathematics department:"Block Partitions of Sequences".

https://www.math.cmu.edu/math/acoseminars/acoseminar.php?SeminarSelect=922


Predrag



From schneide at cs.cmu.edu  Fri Feb  7 17:14:44 2014
From: schneide at cs.cmu.edu (Jeff Schneider)
Date: Fri, 07 Feb 2014 17:14:44 -0500
Subject: [Research] Raja Hafiz Affandi, Monday 2/10, noon, GHC 6115
In-Reply-To: <CAMptXfjrzWMzcXyuCj7WBBSZVKxf598bFtS2Jj9OBttbOs9hQw@mail.gmail.com>
References: <CAMptXfjrzWMzcXyuCj7WBBSZVKxf598bFtS2Jj9OBttbOs9hQw@mail.gmail.com>
Message-ID: <52F55AD4.3020207@cs.cmu.edu>

Hi Everyone,

Please attend the ML Lunch talk by Raja Affandi this Monday at noon.  Raja is 
here to interview for a postdoc position in the Auton Lab.

Jeff.



-------- Original Message --------
Subject: 	[ML Lunch] Raja Hafiz Affandi, Monday 2/10, noon, GHC 6115
Date: 	Wed, 5 Feb 2014 18:21:16 -0500
From: 	Leila Wehbe <lwehbe at cs.cmu.edu>
To: 	learning lunch <learning-lunch-seminar at cs.cmu.edu>



Please join us for ML Lunch Monday 2/10 at noon in GHC 6115.

For more information and to see previous talks please visit our website:
http://www.cs.cmu.edu/~learning/ <http://www.cs.cmu.edu/%7Elearning/>.

Speaker: Raja Hafiz Affandi (UPenn)

Title: Large Scale Inference of Determinantal Point Processes (DPPs)

Abstract:
Determinantal Point Processes (DPPs) are random point processes well-suited for
modelling repulsion. In machine learning and statistics, DPPs are a natural
model for subset selection problems where diversity is desired. For example,
they can be used to select diverse sets of sentences to form document summaries,
or to return relevant but varied text and image search results, or to detect
non-overlapping multiple object trajectories in video. Among many remarkable
properties, they offer tractable algorithms for exact inference, including
computing marginals, computing conditional probabilities, and sampling. In our
recent work, we extended these algorithms to approximately infer non-linear DPPs
defined over a large amount of data, as well as DPPs defined on continuous
spaces using low-rank approximations. We demonstrated the advantages of our
models on several machine learning and statistical tasks: motion capture video
summarization, repulsive mixture modelling and synthesizing diverse human poses.
Given time, I will also briefly touch on our other related works such as
extending DPPs into a temporal process that sequentially select multiple diverse
subsets across time and how we go about learning the parameters of a DPP kernel.
These are joint works with Emily Fox, Ben Taskar and Alex Kulesza.