From nicolas.brodu at numerimoire.net Fri May 8 07:43:09 2009 From: nicolas.brodu at numerimoire.net (Nicolas Brodu) Date: Fri, 8 May 2009 09:43:09 +0200 Subject: [Causality-ML] Next talk: STEMMATOLOGY: causal discovery of relationships between documents Message-ID: <200905080943.10275.nicolas.brodu@numerimoire.net> Dear Causality and Machine Learning group, The next presentation will be on Thursday 14 May, by Patrik Hoyer. He will present us Stemmatology, and discuss the corresponding causality workshop results by Roos et al. 1. Please read the following references: http://www.cs.helsinki.fi/u/ttonteri/casc/stemma.html http://llc.oxfordjournals.org/cgi/reprint/fqp002?ijkey=sf11IHP4UokyTJU&keytype=ref 2. Read the paper, and discuss the results on the list. 3. Connect to the live presentation and ask questions to the lecturer. When: Thursday 14 May, Paris 17h, ET 11h, PT 8h, Tokyo 0h(Friday) URL: http://www.afia-france.org/tiki-index.php?page=GroupeDeLecture090514PH Phone number: +1 (218) 936-7999 Participant code: 665140# See below for tips how to access the conference on the Internet using voice-over-IP. Abstract of the presentation: -------- Stemmatology studies relations among different variants of a document that have been derived from an original text by copying and modifying earlier versions. I will describe the causality workbench STEMMATOLOGY task contributed by Roos et al. This task provides an interesting test case for causal discovery from single observations (objects) as opposed to random variables; as recently discussed by Janzing and Schoelkopf (2008). The problem setting is very general, and also widely applicable to problems in phylogenetics, historical linguistics, and archaeology, among others. -------- If you know of potentially interested speakers or if you wish to present a paper, please send us a message so we can add you in the planning. The planning for the next presentation is maintained at: http://www.afia-france.org/tiki-index.php?page=Groupe+de+lecture Tip for a cheap voice-over-IP connection to the conference: Open an account on nonoh.net (server in Germany) or voipcheap.com (server in US) or another SIP provider near your country. Pay 10 euros for 3 or 4 month unlimited access to nearly all land line phone numbers in the world, including the teleconference number. This means you will be able to spend as much time as you want for free on long-distance and international calls with the basic fee. Use a microphone + headset on your computer, and a program like Twinkle (Linux) or Ekiga (multi-OS) or X-Lite (windows). Set it up with your nonoh.net or voipcheap account (see the help page on these sites). Beware of filtering firewalls. Then call our teleconference number from your computer. Note: We are not affiliated to any of these groups. We provide this information in the hope it is useful. Best regards, Nicolas Brodu From nicolas.brodu at numerimoire.net Mon May 11 22:44:29 2009 From: nicolas.brodu at numerimoire.net (Nicolas Brodu) Date: Tue, 12 May 2009 00:44:29 +0200 Subject: [Causality-ML] Reminder: STEMMATOLOGY: causal discovery of relationships between documents Message-ID: <200905120044.29651.nicolas.brodu@numerimoire.net> Dear Causality and Machine Learning group, This is a friendly reminder for the next presentation on Thursday 14 May, by Patrik Hoyer. He will present us Stemmatology, and discuss the corresponding causality workshop results by Roos et al. 1. Please read the following references: http://www.cs.helsinki.fi/u/ttonteri/casc/stemma.html http://llc.oxfordjournals.org/cgi/reprint/fqp002?ijkey=sf11IHP4UokyTJU&keytype=ref 2. Read the paper, and discuss the results on the list. 3. Connect to the live presentation and ask questions to the lecturer. When: Thursday 14 May, Paris 17h, ET 11h, PT 8h, Tokyo 0h(Friday) URL: http://www.afia-france.org/tiki-index.php?page=GroupeDeLecture090514PH Phone number: +1 (218) 936-7999 Participant code: 665140# See below for tips how to access the conference on the Internet using voice-over-IP. Abstract of the presentation: -------- Stemmatology studies relations among different variants of a document that have been derived from an original text by copying and modifying earlier versions. I will describe the causality workbench STEMMATOLOGY task contributed by Roos et al. This task provides an interesting test case for causal discovery from single observations (objects) as opposed to random variables; as recently discussed by Janzing and Schoelkopf (2008). The problem setting is very general, and also widely applicable to problems in phylogenetics, historical linguistics, and archaeology, among others. -------- If you know of potentially interested speakers or if you wish to present a paper, please send us a message so we can add you in the planning. The planning for the next presentation is maintained at: http://www.afia-france.org/tiki-index.php?page=Groupe+de+lecture Tip for a cheap voice-over-IP connection to the conference: Open an account on nonoh.net (server in Germany) or voipcheap.com (server in US) or another SIP provider near your country. Pay 10 euros for 3 or 4 month unlimited access to nearly all land line phone numbers in the world, including the teleconference number. This means you will be able to spend as much time as you want for free on long-distance and international calls with the basic fee. Use a microphone + headset on your computer, and a program like Twinkle (Linux) or Ekiga (multi-OS) or X-Lite (windows). Set it up with your nonoh.net or voipcheap account (see the help page on these sites). Beware of filtering firewalls. Then call our teleconference number from your computer. Note: We are not affiliated to any of these groups. We provide this information in the hope it is useful. Best regards, Nicolas Brodu From nicolas.brodu at numerimoire.net Thu May 14 22:10:38 2009 From: nicolas.brodu at numerimoire.net (Nicolas Brodu) Date: Fri, 15 May 2009 00:10:38 +0200 Subject: [Causality-ML] Video replay available (Stemmatology), next talk is an overview of the LOCANET task (Pot-luck challenge, NIPS 2008) Message-ID: <200905150010.38934.nicolas.brodu@numerimoire.net> Dear Causality and Machine Learning group, The video of today's talk by Patrik Hoyer is now available online! http://www.encours.org/causality/stemmatology/replay.html Do not hesitate to contact Patrick or ask on the list if you have questions or if you want to try the task he presented today. The next presentation is by Isabelle Guyon. She will give an overview of the LOCANET task from the Pot-luck challenge and NIPS 2008 workshop: When: Thursday 21 May, Paris 17h, ET 11h, PT 8h, Tokyo 0h(Friday) URL: http://www.afia-france.org/tiki-index.php?page=GroupeDeLecture090521PD Abstract of the next presentation: -------- LOCANET stands for LOcal CAusal NETwork. We designed four datasets for the purpose of benchmarking local causal discovery algorithms. These include two "re-simulated" datasets obtained from data artificially generated by models trained with real data and two datasets including real variables intermixed with artificial variables (called probes). There is no time dependency in the samples. We chose applications in marketing, pharmacology and bio-medicine spanning a high diversity of types of distributions. The datasets were used in two challenges in 2008 organized for the WCCI and NIPS conferences. In this presentation, we will focus on the results of the NIPS 2008 challenge: learn depth 3 networks around a target variable from observational data. We will describe the four datasets, explain the task and the evaluation procedures, and analyze the results. The website of the challenges remains open for post-challenge submissions. -------- If you know of potentially interested speakers or if you wish to present a paper, please send us a message so we can add you in the planning. Best regards, Nicolas Brodu From nicolas.brodu at numerimoire.net Tue May 19 07:39:04 2009 From: nicolas.brodu at numerimoire.net (Nicolas Brodu) Date: Tue, 19 May 2009 09:39:04 +0200 Subject: [Causality-ML] Reminder: overview of the LOCANET task (Pot-luck challenge, NIPS 2008) Message-ID: <200905190939.04474.nicolas.brodu@numerimoire.net> Dear Causality and Machine Learning group, This is a friendly reminder for the next presentation by Isabelle Guyon. She will give an overview of the LOCANET task from the Pot-luck challenge and NIPS 2008 workshop: When: Thursday 21 May, Paris 17h, ET 11h, PT 8h, Tokyo 0h(Friday) URL: http://www.afia-france.org/tiki-index.php?page=GroupeDeLecture090521PD Phone number: +1 (218) 936-7999 Participant code: 665140# See below for tips how to access the conference on the Internet using voice-over-IP. Abstract of the next presentation: -------- LOCANET stands for LOcal CAusal NETwork. We designed four datasets for the purpose of benchmarking local causal discovery algorithms. These include two "re-simulated" datasets obtained from data artificially generated by models trained with real data and two datasets including real variables intermixed with artificial variables (called probes). There is no time dependency in the samples. We chose applications in marketing, pharmacology and bio-medicine spanning a high diversity of types of distributions. The datasets were used in two challenges in 2008 organized for the WCCI and NIPS conferences. In this presentation, we will focus on the results of the NIPS 2008 challenge: learn depth 3 networks around a target variable from observational data. We will describe the four datasets, explain the task and the evaluation procedures, and analyze the results. The website of the challenges remains open for post-challenge submissions. -------- If you know of potentially interested speakers or if you wish to present a paper, please send us a message so we can add you in the planning. The planning for the next presentation is maintained at: http://www.afia-france.org/tiki-index.php?page=Groupe+de+lecture Tip for a cheap voice-over-IP connection to the conference: Open an account on nonoh.net (server in Germany) or voipcheap.com (server in US) or another SIP provider near your country. Pay 10 euros for 3 or 4 month unlimited access to nearly all land line phone numbers in the world, including the teleconference number. This means you will be able to spend as much time as you want for free on long-distance and international calls with the basic fee. Use a microphone + headset on your computer, and a program like Twinkle (Linux) or Ekiga (multi-OS) or X-Lite (windows). Set it up with your nonoh.net or voipcheap account (see the help page on these sites). Beware of filtering firewalls. Then call our teleconference number from your computer. Note: We are not affiliated to any of these groups. We provide this information in the hope it is useful. Best regards, Nicolas Brodu From guyon at clopinet.com Thu May 21 04:52:46 2009 From: guyon at clopinet.com (Isabelle Guyon) Date: Wed, 20 May 2009 21:52:46 -0700 Subject: [Causality-ML] Talk starting Message-ID: <4A14DE1E.3040700@clopinet.com> We are starting an overview of the LOCANET task from the Pot-luck challenge: When: Thursday 21 May, Paris 17h, ET 11h, PT 8h, Tokyo 0h(Friday) URL: http://www.afia-france.org/tiki-index.php?page=GroupeDeLecture090521PD Phone number: +1 (218) 936-7999 Participant code: 665140# Isabelle From nicolas.brodu at numerimoire.net Wed May 27 17:41:12 2009 From: nicolas.brodu at numerimoire.net (Nicolas Brodu) Date: Wed, 27 May 2009 19:41:12 +0200 Subject: [Causality-ML] Tomorrow's talk: Discover Local Causal Network around a Target with a Given Depth Message-ID: <200905271941.12211.nicolas.brodu@numerimoire.net> Dear Causality and Machine Learning group, Tomorrow's presentation by Changzang Wang is about discover the local causal network around a target with a given depth. This talks relates to lask week LOCANET task presentation from the Pot-luck challenge and NIPS 2008 workshop: When: Thursday 28 May, Paris 17h, ET 11h, PT 8h, Tokyo 0h(Friday) URL: http://www.afia-france.org/tiki-index.php?page=GroupeDeLecture090528CW Phone number: +1 (218) 936-7999 Participant code: 665140# See below for tips how to access the conference on the Internet using voice-over-IP. Abstract of the presentation: -------- For a given target node T and a given depth k > 0, we propose an algorithm for discovering a local causal network around the target T which has a radius of length k from T. In our algorithm, we find parents, children and some descendants (PCD) of nodes stepwise away from the target T until all edges within the depth k local network cannot be oriented further. Our algorithm extends the PCD-by-PCD algorithm for prediction with intervention presented in Yin et al. (2008). Our algorithm can construct a depth k local network, has a more efficient stop rule and find PCDs along some but not all pathes away from the target. -------- If you know of potentially interested speakers or if you wish to present a paper, please send us a message so we can add you in the planning. The planning for the next presentation is maintained at: http://www.afia-france.org/tiki-index.php?page=Groupe+de+lecture Tip for a cheap voice-over-IP connection to the conference: Open an account on nonoh.net (server in Germany) or voipcheap.com (server in US) or another SIP provider near your country. Pay 10 euros for 3 or 4 month unlimited access to nearly all land line phone numbers in the world, including the teleconference number. This means you will be able to spend as much time as you want for free on long-distance and international calls with the basic fee. Use a microphone + headset on your computer, and a program like Twinkle (Linux) or Ekiga (multi-OS) or X-Lite (windows). Set it up with your nonoh.net or voipcheap account (see the help page on these sites). Beware of filtering firewalls. Then call our teleconference number from your computer. Note: We are not affiliated to any of these groups. We provide this information in the hope it is useful. Best regards, Nicolas Brodu