From dpelleg+ at cs.cmu.edu Wed Sep 5 04:23:24 2007 From: dpelleg+ at cs.cmu.edu (Dan Pelleg) Date: Wed, 5 Sep 2007 11:23:24 +0300 Subject: [auton-users] NIPS*2007 Workshop on Efficient Machine Learning Message-ID: <20070905082324.GB74913@lawrence.libagent.org> NIPS*2007 Workshop on Efficient Machine Learning Overcoming Computational Bottlenecks in Machine Learning Whistler, Canada, December 7-8, 2007 http://bigml.wikispaces.com/cfp Overview -------- The ever increasing size of available data to be processed by machine learning algorithms has yielded several approaches, from online algorithms to parallel and distributed computing on multi-node clusters. Nevertheless, it is not clear how modern machine learning approaches can either cope with such parallel machineries or take into account strong constraints regarding the available time to handle training and/or test examples. This workshop will explore two alternatives: * modern machine learning approaches that can handle real time processing at train and/or at test time, under strict computational constraints (when the flow of incoming data is continuous and needs to be handled) * modern machine learning approaches that can take advantage of new commodity hardware such as multicore, GPUs, and fast networks. This two-day workshop aims to set the agenda for future advancements by fostering a discussion of new ideas and methods and by demonstrating the potential uses of readily-available solutions. It will bring together both researchers and practitioners to offer their views and experience in applying machine learning to large scale learning. Topics of Interest ------------------ * efficient parallelization of machine learning algorithms and algorithms that make use of new hardware architectures * sub-linear training algorithms for virtually infinite datasets * new online boosting, online kernel, and other efficient non-linear online training algorithms * efficient feature extraction for classification and detection * adapted structures for very large number of features per example * evolving under strict time/space constraints * coarse-to-fine and "focusing" algorithms for detection Submission Procedure -------------------- We encourage the submissions of extended abstract. The suggested abstract length is about 2 pages. The invited speakers will be allocated between 40 and 60 minutes, while the authors of the accepted abstracts will be allocated between 30 and 40 minutes to present their work (to be determined according to submissions). In addition, the abstracts will be available to a broader audience on the dedicated web site. The authors should submit their extended abstract to bigml.nips at gmail.com in pdf. An email confirming the reception of the submission will be sent by the organizers. Important Dates --------------- * Aug 28: Workshop announcement / call for abstracts * Oct 12: Abstract submission deadline * Nov 1: Notification of acceptance * Dec 7 and 8: Workshop Invited Speakers ---------------- * Yali Amit, University of Chicago * Yoshua Bengio, University of Montreal * Michael Burl, NASA JPL * Corinna Cortes, Google * Dennis DeCoste, Microsoft * Don Geman, John Hopkins University * Dan Pelleg and Elad Yom-Tov, IBM Research * Yann LeCun, New York University * Srinivasan Parthasarathy, Ohio State University * Nicol N. Schraudolph, National ICT Australia Organizers ---------- * Samy Bengio, Google * Corinna Cortes, Google * Dennis DeCoste, Microsoft Live Labs * Francois Fleuret, IDIAP Research Institute * Ramesh Natarajan, IBM T.J. Watson Research Lab * Edwin Pednault, IBM T.J. Watson Research Lab * Dan Pelleg, IBM Haifa Research Lab * Elad Yom-Tov, IBM Haifa Research Lab Other Members of the Programme Committee ---------------------------------------- * Yali Amit, University of Chicago * Gilles Blanchard, Fraunhofer Institut FIRST, Berlin * Ronan Collobert, NEC * Yves Grandvalet * IDIAP Research Institute * Jiri Matas, Czech Technical University, Prague * Sam Roweis, Google From dpelleg at cs.cmu.edu Wed Sep 5 01:37:57 2007 From: dpelleg at cs.cmu.edu (Dan Pelleg) Date: Wed, 5 Sep 2007 08:37:57 +0300 Subject: [auton-users] NIPS*2007 Workshop on Efficient Machine Learning Message-ID: <20070905053757.GA74032@lawrence.libagent.org> NIPS*2007 Workshop on Efficient Machine Learning Overcoming Computational Bottlenecks in Machine Learning Whistler, Canada, December 7-8, 2007 http://bigml.wikispaces.com/cfp Overview -------- The ever increasing size of available data to be processed by machine learning algorithms has yielded several approaches, from online algorithms to parallel and distributed computing on multi-node clusters. Nevertheless, it is not clear how modern machine learning approaches can either cope with such parallel machineries or take into account strong constraints regarding the available time to handle training and/or test examples. This workshop will explore two alternatives: * modern machine learning approaches that can handle real time processing at train and/or at test time, under strict computational constraints (when the flow of incoming data is continuous and needs to be handled) * modern machine learning approaches that can take advantage of new commodity hardware such as multicore, GPUs, and fast networks. This two-day workshop aims to set the agenda for future advancements by fostering a discussion of new ideas and methods and by demonstrating the potential uses of readily-available solutions. It will bring together both researchers and practitioners to offer their views and experience in applying machine learning to large scale learning. Topics of Interest ------------------ * efficient parallelization of machine learning algorithms and algorithms that make use of new hardware architectures * sub-linear training algorithms for virtually infinite datasets * new online boosting, online kernel, and other efficient non-linear online training algorithms * efficient feature extraction for classification and detection * adapted structures for very large number of features per example * evolving under strict time/space constraints * coarse-to-fine and "focusing" algorithms for detection Submission Procedure -------------------- We encourage the submissions of extended abstract. The suggested abstract length is about 2 pages. The invited speakers will be allocated between 40 and 60 minutes, while the authors of the accepted abstracts will be allocated between 30 and 40 minutes to present their work (to be determined according to submissions). In addition, the abstracts will be available to a broader audience on the dedicated web site. The authors should submit their extended abstract to bigml.nips at gmail.com in pdf. An email confirming the reception of the submission will be sent by the organizers. Important Dates --------------- * Aug 28: Workshop announcement / call for abstracts * Oct 12: Abstract submission deadline * Nov 1: Notification of acceptance * Dec 7 and 8: Workshop Invited Speakers ---------------- * Yali Amit, University of Chicago * Yoshua Bengio, University of Montreal * Michael Burl, NASA JPL * Corinna Cortes, Google * Dennis DeCoste, Microsoft * Don Geman, John Hopkins University * Dan Pelleg and Elad Yom-Tov, IBM Research * Yann LeCun, New York University * Srinivasan Parthasarathy, Ohio State University * Nicol N. Schraudolph, National ICT Australia Organizers ---------- * Samy Bengio, Google * Corinna Cortes, Google * Dennis DeCoste, Microsoft Live Labs * Francois Fleuret, IDIAP Research Institute * Ramesh Natarajan, IBM T.J. Watson Research Lab * Edwin Pednault, IBM T.J. Watson Research Lab * Dan Pelleg, IBM Haifa Research Lab * Elad Yom-Tov, IBM Haifa Research Lab Other Members of the Programme Committee ---------------------------------------- * Yali Amit, University of Chicago * Gilles Blanchard, Fraunhofer Institut FIRST, Berlin * Ronan Collobert, NEC * Yves Grandvalet * IDIAP Research Institute * Jiri Matas, Czech Technical University, Prague * Sam Roweis, Google From mjbaysek at cs.cmu.edu Sat Sep 15 17:50:03 2007 From: mjbaysek at cs.cmu.edu (Michael J. Baysek) Date: Sat, 15 Sep 2007 17:50:03 -0400 Subject: [auton-users] Auton Lab Data Transfers to Offsite Networks: Please Read Message-ID: <46EC538B.3010503@cs.cmu.edu> Hi Lab I want to take a minute to point out that we are under some limitations when transferring data across the Internet. In order to maintain quality of service to the entire university, restrictions on Internet traffic usage is placed on every machine on the network. On Friday, one or more of you transferred in excess of 10 GB of data over the Internet using LOS1. Now, transferring this much inside of CMU is perfectly fine, but outside of CMU, I get a notice. Too many notices and a machines network connection can be suspended indefinitely by computing services without notice. It is for this reason that I ask you - If you have large amounts of data to transfer to (or from) offsite networks, please contact me in advance so that I can arrange for this to be done in a way that can circumvent the bandwidth limitations. I have prearranged with computing services to allow certain hosts to transfer larger amounts of data. If you need to make a large (on the order of GB) data transfer , please contact me, and I can set you up with a way to do it without affecting the access for everyone. Thanks, Mike -------- Original Message -------- Subject: NN48385 - Notification: Network Bandwidth Limit Exceeded - LOS1.AUTON.CS.CMU.EDU (fwd) Date: Sat, 15 Sep 2007 09:48:06 -0400 (EDT) From: Gary Aranyos To: CC: ---------- Forwarded message ---------- Date: Sat, 15 Sep 2007 09:08:55 -0400 From: abuse at andrew.cmu.edu To: zack at andrew.cmu.edu, mp23 at andrew.cmu.edu, cs1y at andrew.cmu.edu, gward at andrew.cmu.edu, aranyos at andrew.cmu.edu, jhutz at andrew.cmu.edu, cpc at andrew.cmu.edu Cc: acs+ng.project.epidemic.outgoing at andrew.cmu.edu Subject: NN48385 - Notification: Network Bandwidth Limit Exceeded - LOS1.AUTON.CS.CMU.EDU TO: FROM: Carnegie Mellon Computing Services Computing and Network Security SUBJECT: Notification: Network Bandwidth Limit Exceeded MACHINE: Machine: Machine: IP 128.2.204.47, MAC , Hostname LOS1.AUTON.CS.CMU.EDU You are receiving this message because the machine listed above has used an amount of network bandwidth in excess of the current guidelines. Current guidelines allow for a single machine to generate two (2) Gigabytes of traffic per day on the wired network, either inbound (from the internet) or outbound (to the internet). You can read the guidelines at: The following shows the total daily usage of network bandwidth for this machine over the past five (5) days: Daily Rates: Inbound Outbound Date Usage Usage 2007-09-10 0.002GB 0.004GB* 2007-09-11 0.001GB 0.003GB* 2007-09-13 0.001GB 0.002GB* 2007-09-14 11.087GB* 0.289GB Five Day Average: 2.219 Note that the Five Day Average counts the greater value of inbound or outbound traffic each day. We request that you reduce your bandwidth use immediately to under two (2) Gigabytes per day. Since this is your first notification regarding your bandwidth usage, Computing Services will send you a warning if your machine uses more than two (2) Gigabytes of either inbound or outbound bandwidth in a single day. If you continue to exceed the guidelines after that warning, we will send one additional warning before we will be forced to suspend network access for this machine to ensure that all users have reasonable access to the network. Computing Services enforces bandwidth quotas to ensure that a small number of machines do not degrade the network service by consuming an excessive amount of available bandwidth. If you have a demonstrated need for using bandwidth that exceeds the guideline limit, you can request an exemption as described later in this message. --------------------------------------- Why would I be using so much bandwidth? In many cases systems generate excessive bandwidth without the owners knowlege or intent. Services such as peer-to-peer file sharing programs (kazaa, gnutella, bittorrent, and others) often serve content by default which can contribute to this problem. It is also possible that the machine is infected with a worm which is generating traffic, or that the machine has been compromised and is being used as a server or to launch attacks. Downloading or uploading high volumes of software, documents, or programs can also cause excessive usage. --------------------------------------- What if I have a need to use this much bandwidth? If your use of network bandwidth is the result of a demonstrated need that supports the educational and research goals of the university, you can fill out the form on the following web page to request an exemption: Requesting an exemption does not guarantee that you will receive one. If you do not receive an exemption, you will need to reduce your network bandwidth usage immediately to under two (2) Gigabytes per day. Failure to reduce the bandwidth usage for systems which are not granted an exemption may result in the system being blocked from using the campus network. --------------------------------------- What should I do if I don't recognize the machine you're talking about? You have received this message because you are listed as an owner of the machine in NetReg () or because you are a member of an administrative group with responsibility for the machine. The listed owner(s) of this system are: The listed administrator(s) of this machine: If this machine does not belong to you, try contacting one of the administrators listed above. If you are one of the administrators listed, please contact abuse at andrew.cmu.edu. Thank you, Computer and Network Security Team Computing Services From mjbaysek at cs.cmu.edu Wed Sep 19 16:22:06 2007 From: mjbaysek at cs.cmu.edu (Michael J. Baysek) Date: Wed, 19 Sep 2007 16:22:06 -0400 Subject: [auton-users] System Back up and running Message-ID: <46F184EE.9010504@cs.cmu.edu> Hi Lab Today around noon, we had another system failure in our main file server. We are again running from the backup file server while I determine the problem with the primary server. During this time, WebCVS and SVNWeb services at https://www.autonlab.org/cvs/ and https://www.autonlab.org/svn/ , respectively will not be available. I will keep you informed of any important updates or changes. Please feel free to call me if you think there are any other problems. Mike -- Michael J. Baysek, Systems Analyst Carnegie Mellon University - Auton Lab www.cmu.edu - www.autonlab.org 412-268-8939 Office 412-596-1618 Cell