dense outliers

JoCG financial statement

2012-07-23T21:23:00.000+10:00

Pat put together the financial statements for JoCG.

"In order to help others understand what the real cost of running an open access journal are, JoCG has released their financial statements for 2010, 2011, and 2012 (YTD)."

JoCG renewing editorial board

2012-06-01T14:55:00.000+10:00

JoCG has a policy to renew the editorial board regularly. This year the editorial board got seven new members while eight will be leaving during the year. The new members are: Joachim Giesen, Yoshio Okamoto, Herman Haverkort, Dominique Attali, John Iacono, Janos Pach, Uli Wagner. I'm sure they'll do a great job for JoCG!

COCOON 2012: 1st call for papers

2012-01-24T10:10:00.001+11:00

CALL FOR PAPERS: COCOON 2012

The 18th Annual International Computing and Combinatorics Conference
20-22 August 2012, Sydney, Australia.
http://cocoon.it.usyd.edu.au/

Original research papers in the areas of algorithms, theory of computation,
computational complexity, and combinatorics related to computing are
solicited. In addition to theoretical results, we are interested in submissions
that report on experimental and applied research of general algorithmic
interest. The conference proceedings will be published by Springer-Verlag
in the Lecture Notes in Computer Science series. Typical, but not exclusive,
topics of interest include:
- Algorithms and Data Structures
- Algorithmic Game Theory
- Approximation and online algorithms
- Automata, Languages, Logic, and Computability
- Combinatorics Related to Algorithms and Complexity
- Complexity Theory
- Computational Learning Theory and Knowledge Discovery
- Cryptography, Reliability and Security, and Database Theory
- Computational Biology and Bioinformatics
- Computational Algebra, Geometry, and Topology
- Graph Drawing and Information Visualization
- Graph Theory, Communication Networks, and Optimization
- Parallel and Distributed Computing

Paper Submission
Contributors are invited to submit a full paper with a maximum of 12 pages
in the Springer LNCS format including title, abstract and references. An
appendix, beyond the 12 pages, may be added and may be read at the reviewers'
discretion. Please format your paper in the Springer Lecture Notes style
according to the LNCS Author Instructions. Submitted papers must describe work not
previously published. They must not be submitted simultaneously to another conference with
refereed proceedings or to a journal. Only electronic submission (pdf) will be allowed.

Publication
The conference proceedings will be published by Springer-Verlag in the Lecture Notes in Computer
Science (LNCS) series. After the conference, selected papers will be published in special issues.

Awards
A best paper award and a best student paper award will be given. A paper is eligible for the best
student paper award if all authors are full-time students at the time of submission.

Student grants
There will be a number of student travel awards consisting in free registration. These awards will
be granted on the merits of the applicants. Priority will be given to students presenting a paper
at the meeting.

Invited Speakers
Kamal Jain
Joseph Mitchell, Stony Brook

Important Dates
Submission deadline February 18, 2012 (by 11:59:59pm CST)
Notification of acceptance April 18, 2012
Final version due May 12, 2012

Contact: cocoon2012@it.usyd.edu.au

Open position in computer systems at USyd

2011-09-20T23:05:00.002+10:00

Our school is currently advertising a (tenure-tracked) position at the Lecturer (Level B)/Senior Lecturer level (Level C) in computer systems.

More details at: http://tinyurl.com/3tzfwqw

JoCG indexed

2011-03-21T17:50:00.002+11:00

JoCG is now included the Directory of Open Access Journals and MathSciNet (thanks to David Wood).

Which other organisations should we aim to get it indexed by?

Thore on theory on TV

2011-01-13T20:07:00.004+11:00

I just spent the last 22:06 minutes watching a great little film with Thore Husfeldt where he talks about algorithms, in particular Google’s PageRank.

Thore wrote a post about the work that you can find here.

Unfortunately it's only available in Danish (a dialect of Swedish).

Theoretical computer science day in Sydney

2010-12-09T09:58:00.001+11:00

Theoretical computer science day in Sydney
January 28, 2011
New Law School Seminar 020, University of Sydney

CALL FOR PARTICIPATION
The theoretical computer science day will take place in Sydney on
January 28, 2011. The aim of this one-day event is to bring together
researchers interested in all aspects of theoretical computer science,
and discuss current trends and problems in the field, as well as
future directions. We would like to invite researchers in theoretical
computer science to participate in this event. If you are interested
in participating, please register as soon as possible for the event
through the website mentioned below. Registration is open to everyone
and is free.

More information is available on the web at
http://sact.it.usyd.edu.au/TDS, including information about the
invited presentations, the location and registration.

The list of invited speakers includes:
Otfried Cheong, KAIST, Korea
Catherine Greenhill, UNSW
Mohammad Mahdian, Yahoo! research
Bernard Mans, Macquarie University
Toby Walsh, NICTA
David Wood, University of Melbourne

ORGANISATION
Joachim Gudmundsson, NICTA, joachim.gudmundsson@nicta.com.au
Julian Mestre, University of Sydney, mestre@it.usyd.edu.au
Taso Viglas, University of Sydney, taso.viglas@sydney.edu.au

Another great JoCG article

2010-11-30T18:04:00.001+11:00

Computing the Maximum Detour of a Plane Geometric Graph in Subquadratic Time
by Christian Wulff-Nilsen

Abstract:
Let G be a plane geometric graph where each edge is a line segment. We consider the problem of computing the maximum detour of G, defined as the maximum over all pairs of distinct points (vertices as well as interior points of edges) p and q of G of the ratio between the distance between p and q in G and the Euclidean distance ||pq||2. The fastest known algorithm for this problem has Θ(n2) running time where n is the number of vertices. We show how to obtain O(n3/2(log n)3) expected running time. We also show that if G has bounded treewidth, its maximum detour can be computed in O(n(log n)3) expected time.

JoCG paper

2010-11-21T19:43:00.001+11:00

New paper published in JoCG:

COMPUTING MULTIDIMENSIONAL PERSISTENCE
Gunnar Carlsson, Gurjeet Singh, Afra J. Zomorodian

ABSTRACT
The theory of multidimensional persistence captures the topology of a multifiltration - a multiparameter family of increasing spaces. Multifiltrations arise naturally in the topological analysis of scientific data. In this paper, we give a polynomial time algorithm for computing multidimensional persistence. We recast this computation as a problem within computational commutative algebra and utilize algorithms from this area to solve it. While the resulting problem is EXPSPACE-complete and the standard algorithms take doubly-exponential time, we exploit the structure inherent withing multifiltrations to yield practical algorithms. We implement all algorithms in the paper and provide statistical experiments to demonstrate their feasibility.

Call for papers!

2010-10-19T11:57:00.003+11:00

Some recent CG related cfp:

- Symposium on Computational Geometry (SoCG'11) will be held on June 13-15 in Paris, France. Deadline: Dec 1

There is also a CG Learning Summer School (June 9-11) as a satellite event.

- European Workshop on Computational Geometry (EuroCG'11) will be held on March 28-30, 2011, in Morschach, Switzerland. Deadline: Jan 9

- Conference on Geosensor Networks (GSN'11) will be held on July 11-13, 2011, in Melbourne, Australia. Deadline: Mar 4

Some SoCG'10 papers

2010-06-30T19:58:00.005+10:00

SoCG is over for this year. In general I thought that the quality of the accepted papers was very high, and I really enjoyed many of the presentations and the results. There has already been numerous reports from SoCG (Sorelle I, II and III, David, Sariel and Suresh), and below I'm just going to briefly mention three of the ones that I found the most interesting - in order of presentation.

Timothy Chan - Optimal Partition Trees (Paper).
Timothy improves on a result from 1992 by Matousek. He presents a simpler construction with a better preprocessing time. There are already numerous blog comments on it (Sariel and Sorelle).

Joe Mitchell - A constant factor approximation algorithm for TSP with neighbourhoods in the plane (paper).
This is a problem I've been working on back and forth for a long time. Given a collection of objects (usually polygons) in the plane, called neighbourhoods, find the shortest tour that visits all neighbourhoods. In 2002 we presented a constant factor approximation algorithm for fat disjoint objects of arbitrary size. This was the first constant factor approximation algorithm for arbitrary size objects. Joe has several results on the topic and in his SoCG paper he presents the first constant-factor approximation algorithm for TSPN on an arbitrary set of disjoint, connected neighbourhoods in the plane. This has been an open problem since the mid-90's and Joe's paper is a big step forward. Of course, ultimately we would like a constant-factor approximation for the general case, not necessarily disjoint neighbourhoods. He has several interesting general observations in his paper. One is that skinny neighbourhoods are not longer a big problem. Using “directional hulls”, which are fat, for an appropriate choice of one of two coordinate systems. Then one can partition the regions into two sets (“blue” regions and “red” regions). Separately find approximating tours of each set, and then use the Combination Lemma (Arkin and Hassin) to combine into a tour of the entire set.

Anne Driemel, Sariel Har-Peled and Carola Wenk - Approximating the Frechet distance for realistic curves in near linear time (paper).
This is probably my favourite paper at SoCG this year. It's an important problem (imho) and it's great to finally see a subquadratic time algorithm with reasonable assumptions. The aim is to calculate the Frechet distance between two curves. They introduce a notion of k-bounded curvature which means that in a ball of radius r the length of the trajectory inside that ball is bounded by k*r (this family of curves is closed under simplification, a property that is very useful in practice). This sounds like a reasonable assumption for many applications (although not for trajectories generated from football players). They present a (1+eps)-approximation algorithm to compute the Frechet distance between two curves with k-bounded curvature and prove that this can be done in near-linear time. The key idea is that after the curves have been simplified the complexity of the reachable free space is linear for fixed k and epsilon (the free space diagram is the standard tool for computing the Frechet distance between curves but has quadratic complexity).

New position at the university of sydney (algorithmic machine learning)

2010-06-29T21:17:00.000+10:00

We have a new position in "algorithmic machine learning" at the university of sydney.
The deadline for submitting an application is 12-august-2010.

The position is at the level of "lecturer" which is the same as an assistant professor, or "senior lecturer" which is again the same as an assistant professor but usually slightly older, with some gray hair, possibly kids, and a mortgage.

JoCG new article

2010-06-08T22:45:00.003+10:00

The second scientific article in JoCG was just published. It's an expository of visibility and blockers by Attila Por and David Wood. Enjoy!

CATS 2011 call for papers (the australian theory conference)

2010-05-25T12:01:00.000+10:00

CATS 2011 -- Computing: The Australasian Theory Symposium
Perth, Australia, January 17-20, 2011
http://cats.it.usyd.edu.au/

The 17th Computing: The Australasian Theory Symposium (CATS) will be held in Perth, Australia, January 17-20, 2011. CATS is an annual conference held in the Australia-New Zealand region, dedicated to theoretical computer science.

Authors are invited to submit papers that present original and unpublished research on topics including (but not limited to) the following areas: algorithms and data structures, complexity theory, graph theory, graph algorithms and combinatorics, semantics of programming languages, approximation and randomized algorithms, combinatorial optimization, formal program specification and transformation, computational geometry, algorithmic game theory, computational biology, logic and type systems, and computability.

Deadlines and other dates:
Paper submission deadline:Monday August 16, 2010
Acceptance notification:Monday October 4, 2010
Final version of accepted papers due:Monday November 5, 2010
Early registration: Monday December 6, 2010
Conference dates: January 17-20, 2011

The proceedings of this event will be published by the Australian Computer Society (ACS) in the CRPIT Series (http://crpit.com/), and will also appear in the ACM digital library.

CATS 2011 is part of the Australasian Computer Science Week (ACSW), an international annual conference event, supported by the Computing Research and Education Association (CORE) in Australia. ACSW 2011 is hosted by Curtin University in Perth, Australia.

For more information please visit http://cats.it.usyd.edu.au/
Contact: taso viglas (aviglas+cats2011@gmail.com)

Back in the office

2010-05-11T17:02:00.003+10:00

Today is my first day back at work after three months daddy leave. It's been great (I highly recommend it!) but it's also good to be back. I feel pretty energetic and motivated, but I'm sure it's just temporary.

As part of the new start I refurbished my homepage (can you guess what the banner depicts?), I bought two new sweaters and I just printed (I really should get an ereader) 6 papers that I have to review. Yes, I'm ready for work!

SoCG paper

2010-02-15T11:08:00.001+11:00

As Sariel already noted the SoCG notification is out (couldn't find a list of accepted papers).

I was lucky to get a paper accepted. The paper is on visibility testing in the plane and it's together with Pat Morin.

In the paper we consider query versions of visibility testing and visibility counting. That is, let S be a set of n disjoint line segments in the plane and let s be an element of S. Visibility testing is to preprocess S so that we can quickly determine if s is visible from a query point q. The counting version involves preprocessing S so that one can quickly estimate the number of segments in S visible from a query point q.

The key idea is that one can cover the visibility polygon of a segment with only an expected linear number of triangles (even though its complexity may be quartic), and for all the segments with a quadratic number of triangles, such that given a query point q the number of triangles stabbed is a 2-approximation of the number of segments q can see. Pretty nifty, eh?

JoCG first publications!

2010-02-12T09:34:00.003+11:00

I'm sure some of you are wondering what's happening with JoCG (Journal of Computational Geometry). The first papers were published yesterday!

The first paper of JoCG is a Welcome from the Editors-in-Chief by Ken Clarkson and Günter Rote. And the first contribution to be published is: Happy Endings for Flip Graphs by David Eppstein.

So far JoCG has received 21 submissions. Of these, 1 has been accepted and published, 4 have been accepted pending (various levels of) revisions, 8 have been rejected, and the remaining 8 papers are still being reviewed.

All in all, I think it's a good start of the journal. Hopefully it will get even more submissions now when papers are getting published.

Computational geometry - main stream convex optimisation?

2010-01-20T17:42:00.003+11:00

As Tasos pointed out there's an open position in "Large Scale Convex Optimization" at University of Sydney.

For that reason I would be very interested to hear if anyone has any ideas on how one can argue that computational geometry is actually "Large Scale Convex Optimization", or even that it's remotely related to it.

Really good ideas will be awarded with beer (non-alcoholic of course).

Position at the University of Sydney

2010-01-15T11:53:00.002+11:00

There is a faculty position at the University of Sydney (School of IT). The full ad can be found here.

The position is in the area of optimization and algorithms.

Applications need to be submitted by January 27. The position is at the Lecturer/Senior Lecturer level which is what an assistant professor is called in australia (senior lecturer can also be considered as an early associate professor level appointment).

--taso

postdoc position at the university of sydney

2009-06-19T12:35:00.002+10:00

The University of Sydney is offering 10 postdoctoral fellowships in 2010.
short summary:

up to 3 years,
for phds awarded after 1-Jan-2004,
research grant up to 25,000 aud, plus some relocation expenses
salary around 72,000 aud
(i think... that's USyd level A8 according to this pdf, linked from this page)
fellowship might involve up to 6h per week of "other duties" (teaching)
research proposal for the application is 4 pages

Applicants need to contact the Head of the proposed school/department before august 7 (for computer science that's sanjay chawla at hos@it.usyd.edu.au).
more details in this pdf.

Copying the announcement from the University of Sydney website:

Postdoctoral Research Fellowships

CLOSING DATE TO RESEARCH OFFICE: 11 September 2009

Applicants must contact the Head of the host School/Department before 7 August 2009.

The University of Sydney Postdoctoral Fellowships were established in 1996 to support excellence in full-time research undertaken in any Department or School at the University. The University will be offering up to 10 new Fellowships in 2010. The Fellowships are extremely prestigious and highly competitive internationally in line with equivalent externally funded fellowships. Applicants must have an outstanding track record relative to opportunity in order to be short-listed. Successful applicants are expected to be based full-time at the University for the duration of the Fellowship and must not hold a concurrent paid appointment.

Eligibility

Strong preference will be given to applicants seeking to join the University from another organisation in Australia or from overseas. Applicants must have a PhD award dated no earlier than 1 January 2004.. Applicants with a Phd award dated later than 31 December 2008 are extremely unlikely to be competitive and should talk to the host Head of School to assess competitiveness before applying.
Applicants with a PhD awarded by the University of Sydney within the timeframe specified above may apply if they have held a position with another organisation subsequent to the award of their PhD. Applicants currently employed at the University of Sydney or other affiliated institutions (including but not limited to medical institutes) who commenced such employment after the award of their PhD AND on or after 1 July 2008 are eligible to apply.
Applicants wishing to clarify eligibility must submit a written request by 24 July 2009 to the Research Office. research@usyd.edu.auresearch[at]usyd.edu.au

Assessment Criteria

Excellence will be the primary criterion, both in terms of the project and the researcher. Equal weight will be given to the quality of the project and the track record of the applicant relative to opportunity. The alignment of the proposed research with existing activity and the environment in the host Department/School will also be an important consideration.

Journal of Computational Geometry

2009-06-11T16:43:00.001+10:00

The Journal of Computational Geometry (jocg.org) is now accepting submissions.

Scope and Focus
The Journal of Computational Geometry (JoCG) is an international open access electronic journal devoted to original research of the highest quality in all aspects of computational geometry. JoCG publishes papers on the design and analysis of geometric algorithms, the complexity of geometric problems, experimental work on geometric algorithms, applications of computational geometry, and topics at the intersection of geometry and algorithms. Topics include metric space embeddings, graph drawing, computational topology, topological learning, meshing, compressed sensing, manifold learning, computer-aided design, discrete geometry, and combinatorial geometry. Outstanding survey papers in the area are also considered.

Editors-in-Chief
Ken Clarkson, IBM, United States
Günter Rote, Freie Universität Berlin

Editorial Board
Hee-Kap Ahn, Postech, Korea, Republic of
Oswin Aichholzer, Graz University of Technology, Austria
Nancy M. Amato, Texas A&M University, United States
Lars Arge, University of Aarhus, Denmark
Boris Aronov, Polytechnic University, United States
Mark de Berg, TU Eindhoven, Netherlands
Jean-Daniel Boissonnat, INRIA Sophia-Antipolis, France
Peter Brass, City College of New York, United States
Sergio Cabello, University of Ljubljana, Slovenia
Bernard Chazelle, Princeton University, United States
Otfried Cheong, KAIST, Korea, Republic Of
Ken Clarkson, IBM, United States
Tamal K. Dey, The Ohio State University, United States
Vida Dujmovic, Carleton University, Canada
Jeff Erickson, University of Illinois, Urbana-Champaign, United States
Hazel Everett, Université Nancy, France
Xavier Goaoc, INRIA, France
Anupam Gupta, Carnegie Mellon University, United States
Dan Halperin, Tel Aviv University, Israel
John Hershberger, Mentor Graphics, United States
Ferran Hurtado, Universitat Politècnica de Catalunya, Spain
Piotr Indyk, MIT, United States
Marc van Kreveld, Utrecht University, Netherlands
Stefan Langerman, Université Libre de Bruxelles
Joseph S. B. Mitchell, Stony Brook University, United States
Günter Rote, Freie Universität Berlin
Christian Sohler, TU Dortmund, Germany
Takeshi Tokuyama, Tohoku University, Japan
Jan Vahrenhold, Technische Universität Dortmund, Germany
Yusu Wang, The Ohio State University, United States
David R. Wood, The University of Melbourne, Australia

Managing Editors
Joachim Gudmundsson, NICTA, Australia
Pat Morin, Carleton University, Canada

CATS 2010

2009-06-02T17:55:00.002+10:00

Call for Papers
CATS 2010 -- Computing: The Australasian Theory Symposium
Brisbane, Australia, January 2010
http://cats.it.usyd.edu.au/

CATS (Computing: The Australasian Theory Symposium) is an annual conference dedicated to theoretical computer science. In 2010, the 16th Computing: The Australasian Theory Symposium will be held in Brisbane, Australia, January 18-21, 2010.

Authors are invited to submit papers that present original and unpublished research on topics including (but not limited to) the following areas: algorithms and data structures, complexity theory, graph theory, graph algorithms and combinatorics, semantics of programming languages, approximation and randomized algorithms, combinatorial optimization, formal program specification and transformation, computational geometry, algorithmic game theory, computational biology, logic and type systems, computability and new paradigms of computation.

Deadlines and other dates:
Paper submission deadline: Monday August 17, 2009
Acceptance notification: Monday October 5, 2009
Final version of accepted papers due: Monday November 2, 2009
Early registration: Monday December 7, 2009
Conference dates: January 18-21, 2010

The proceedings of this event will be published by the Australian Computer Society (ACS) in the CRPIT Series (http://crpit.com/), and will also appear in the ACM digital library.

CATS 2010 is part of the Australasian Computer Society Week (ACSW), an international annual conference event, supported by the Computing Research and Education Association (CORE) in Australia. ACSW 2010 is hosted by the School of Information Technology at the Queensland University of Technology (QUT) in Brisbane, Australia, in January 2010.

For more information about CATS please visit http://cats.it.usyd.edu.au/
Contact: cats2010@easychair.org

The birth of a new CG journal

2009-05-30T09:33:00.002+10:00

So we finally decided to start a new journal - Journal of Computational Geometry. After a lot of doubts and discussions back and forth we decided to send out invitations to colleagues asking them to join the editorial board. The responds has been amazingly positive. Even though a few declined they were all very supportive. The, almost finalized, editorial board can be found on the journal webpage (which is still under construction).

We hope to be ready to accept paper in a couple of weeks. More details to follow in a later post.

Should we start a free CG journal?

2009-02-16T10:36:00.003+11:00

After a bit of work we believe we have solved most of the practical problems that have to be taken care of before starting a free journal. This is probably the easy part. Now we have to decide if it is a good idea or not.

The aim is to have a high quality journal for the CG community that is run by the CG community and free to everyone (really free, no cost to publish and no cost to access). Obviously such a journal needs the support of the CG community to be successful. The work should be shared among the community, i.e., the editorial board and editorial manager(s) should be replaced regularly.

Starting it would require a strong commitment from everyone involved (including the editorial board) for the first few years. So do we want a free CG journal? Please let us know what you think by commenting and/or filling in the poll.

You can select "Yes", "No" or "No, but someone else should". The third option is there to capture anyone that is positive to the idea but believes that he or someone else is better suited to do it. We're doing this with the aim to get a free journal in the community and if someone else wants to do it we would be happy to give our support.