FSMNLP. 11th International Conference on Finite-State Methods and Natural Language Processing. July 15-17 2013
St Andrew's, Scotland.
Monday July 15th. A convexity-based generalization of Viterbi for non-deterministic weighted automata
, Marc Dymetman
6th International Conference on Communities and Technologies; 29 June - 02 July 2013; Munich, Germany
XRCE is co-organising the Workshop on Large-Scale Idea Management and Deliberation Systems
.that will take place at Communities and Technologies 201
3: Saturday, June 29th, 2013: Contact: Gregorio Convertino.
XRCE 20th anniversary distinguished speaker series: Jean Ponce, Professor, Director Département d'Informatique de Ecole Normale Supérieure , Paris, France
As part of the XRCE's 20th anniversary events
the centre is running a series of distinguished speaker seminars
Seminar: Modelling visual recognition
; July 9th, 2013 at 11:00 AM.
***For those who missed this seminar the recording will be available week of 15th July ***
CVPR 2013, Portland, Oregon, USA
June 23rd -28th, 2013
Sunday June 23rd. Full day tutorial on Large-Scale Visual Recognition
at CVPR 2013. Florent Perronnin
(XRCE) (primary organizer) Ondrej Chum
(Czech Technical University), Zaid Harchaoui
(INRIA), Herve Jegou
(INRIA), Marc'Aurelio Ranzato
(Google) and Andrea Vedaldi
(Oxford University). More info here
Friday, June 28th.
Second workshop on Fine-Grained Visual Categorization (FGVC) Florent Perronnin
(UC Berkeley), Steve Branson
(Caltech), Neeraj Kumar
(Univ. of Washington) and Anelia Angelova
(NEC Labs). More info here
Friday, June 28th. First workshop on Visual Analysis Beyond Semantics
organized by Luca Marchesotti
Information on Computer Vision research
across the Xerox Innovation Group
XRCE at CERN Workshop on Innovations in Scholarly Communication (OAI8)
University of Geneva, Switzerland, 19-21 June 2013
Seminar: Toward Big Data Analytics, Anton Heijs. CTO & Founder Treparel, June 27 11am.
With the trend towards more data driven decision making, based on the growing amount of available data and technology to process large data sets efficiently, there is a need for flexible data analysis approaches. In the presentation we will discuss how Treparel started using machine learning and visualization to provide quick insight in table, text and image data sets and focused on text analytics especially on classification and clustering of patents and research literature. With this basis we discuss how the KMX technology is extended to help the data scientist to setup analytic pipelines locally and use a sample of the data to test and optimize the analysis before using scalable compute resources like hadoop to process. The benefits of this approach where a small number of information creators (data scientists) present rich interactive information to a large group of information consumers will also be discussed.
Berlin High-level Dialogue on Implementing Rio+20 Decisions on Sustainable Cities and Urban Transport
19- 21 June 2013, Berlin, Germany
and the United Nations Department of Economic and Social Affairs (UN DESA)
have organised this event a year after the Rio +20
conference on sustainable transport
. XRCE is joining in the discussion.
Seminar: "Human recognition using biometrics"; Monday, June 17, 2013 at 11:00 AM
Speaker: Arun Ross
, assistant professor at Michigan State University
, East Lansing, MI, U.S.A.
Biometrics is the science of recognizing humans based on the physical or behavioral attributes of an individual. Read more
International Conference on Artificial Intelligence and Law (ICAIL) 2013; Rome, Italy. June 10 - 14, 2013
Workshop on Standards for Using Predictive Coding, Machine Learning, and Other Advanced Search and Review Methods in E-Discovery (DESI V Workshop)
. Soft Labeling for Multi-Pass Document Review.
Jianlin Cheng, Amanda Jones, Caroline Privault, Jean-Michel Renders
Seminar: "Spectral learning of weighted automata"; June 6th, 2013 at 11:00 AM
Speaker: Franco Luque
, teaching assistant at Universidad Nacional de Córdoba
, Córdoba, Argentina:
In this talk, I will present a derivation of the spectral method for learning Weighted Finite Automata (WFA) that put emphasis on providing intuitions on the inner workings of the method and does not assume a strong background in formal algebraic methods. Read more