Profile

  • 2014-: Associate Professor, Univ. Rennes 1, IRISA, Team Linkmedia
  • 2013-2014: Post doctoral researcher at FEMTO-ST, Besançon on time series mining for prognostics
  • 2012-2013: Assistant Professor at AgroCampus Ouest, Rennes, France, on time series mining
  • 2011-2012: Post doctoral researcher at INESC, Porto, Portugal, on machine learning applied to anomaly detection in network enterprise data
  • 2008-2010: Teaching assistant at the Université de Rennes 1 in computer science and signal processing
  • 2005-2008: PhD Thesis in the TEMICS project (IRISA), on joint source/channel coding for wireless transmission
  • 2005: Master Thesis in the TEMICS project (IRISA), on soft decoding of variable length codes
  • 2004-2005: Master Degree in Signal and Image Processing at the Université de Rennes 1
  • 2004: One semester of courses in the Bachelor of Electrical Engineering, University of Newcastle, Australia
  • 2002-2005: Engineering Degree at the Ecole Nationale Supérieure des Télécommunications de Bretagne

Current research interests

  • Motif extraction in multimedia data
  • Data mining, particularly time series mining
  • Motif extraction in time series
  • Machine learning

Some past research interests

  • Source coding/decoding, joint source and channel coding
  • Distributed arithmetic coding

Publications

Journal papers





  • A. Bailly,S. Malinowski, R. Tavenard, T. Guyet and L. Chapel Dense Bag-of-Temporal-SIFT-Words for time series classification Lecture Notes in Artificial Intelligence - 2016




  • J. Ben Ali, B. Chebel-Morello, L. Saidi, S. Malinowski and F. Fnaiech Accurate bearing remaining useful life prediction based on Weibull distribution and artificial neural network Elsevier Mechanical Systems and Signal Processing - 2015




  • S. Malinowski, B. Chebel-Morello and Noureddine Zerhouni Remaining useful life estimation based on discriminating shapelet extraction Elsevier Reliability Engineering and System Safety Journal - 2015




  • B. Chebel-Morello, S. Malinowski and Noureddine Zerhouni Feature selection for fault detection systems: application to the Tennessee Eastman Process Springer Applied Intelligence Journal - 2015
  • pdf-icon A.K. Marnerides, S. Malinowski, R. Morla and H.S. Kim Fault diagnosis in DSL networks using support vector machines Elsevier Computer Communications Journal (COMCOM) Journal - 2015
  • pdf-icon A. Aubert, R. Tavenard, R. Emonet, A. de Lavenne, S. Malinowski, T. Guyet, R. Quiniou, J-M. Odobez, P. Merot and C. Gascuel Clustering flood events from water quality time-series using Latent Dirichlet Allocation model Water Resources Research (WRR) - Wiley - 2013
  • pdf-icon S. Malinowski, X. Artigas, C. Guillemot and L. Torres Distributed coding using punctured quasi-arithmetic codes for memory and memoryless sources IEEE Transactions on Signal Processing, Oct. 2009, Vol. 57, Number 10
  • pdf-icon S. Malinowski, H. Jégou and C. Guillemot Computation of posterior marginals on aggregated state models for soft source decoding IEEE Transactions on Communications, April 2009, Vol. 57, Number 4
  • pdf-icon S. Malinowski, H. Jégou and C. Guillemot Error recovery properties and soft decoding of quasi-arithmetic codes EURASIP Journal on Advances in Signal Processing, January 2008, Volume 2008, Article ID 752840
  • pdf-icon S. Malinowski, H. Jégou and C. Guillemot Synchronization recovery and state model reduction for soft decoding of variable length codes IEEE Transactions on Information Theory, January 2007, pages 368-377

International conferences and workshops





  • R. Tavenard and S. Malinowski Cost-Aware Early Classification of Time Series in European Conference on Machine Learning and Principles and Practice of Knowledge Discovery (ECML/PKDD) - 2016




  • A. Le Guennec , S. Malinowski and R. Tavenard Data Augmentation for Time Series Classification using Convolutional Neural Networks in ECML/PKDD Workshop on Advanced Analytics and Learnign on Temporal Data - 2016




  • A. Bailly,S. Malinowski, R. Tavenard, T. Guyet and L. Chapel Bag-of-Temporal-SIFT-Words for time series classification ECML/PKDD Workshop on Advanced Analytics and Learnign on Temporal Data - 2015




  • C. Hardy,L. Amsaleg, G. Gravier, S. Malinowski and R. Quiniou Sequential pattern mining on multimedia data ECML/PKDD Workshop on Advanced Analytics and Learnign on Temporal Data - 2015




  • S. Malinowski, B. Morello and N. Zerhouni Shapelet based remaining useful life estimation accepted for IEEE CASE 2014, August 2014, Taipei, Taiwan




  • R. Khelif, S. Malinowski, B. Morello and N. Zerhouni Unsupervised Kernel Regression modeling approach for RUL prediction accepted for PHM Europe, July 2014, Nantes, France




  • R. Khelif, S. Malinowski, B. Morello and N. Zerhouni RUL prediction based on a new similarity-instance based approach Proc. IEEE ISIE 2014, June 2014, Istanbul, Turkey
  • pdf-icon S. Malinowski, T. Guyet, R. Quiniou and R. Tavenard 1d-SAX: a novel symbolic representation for time series Proc. Symposium on Intelligent Data Analysis, October 2013, London, UK
  • pdf-icon A.K. Marnerides, S. Malinowski, R. Morla, M. Rodrigues and H. Kim On the comprehension of DSL SyncTrap events in IPTV Networks Proc. IEEE ISCC'2013, Jul. 7-10, Split, Croatia
  • pdf-icon S. Malinowski and R. Morla A single pass trellis-based algorithm for clustering evolving data streams Proc. DEXA DAWAK'2012, Sep. 3-6, Vienna, Austria




  • A.K. Marnerides, S. Malinowski, R. Morla, M. Rodrigues and H. Kim Towards the Improvement of Diagnostic Metrics: Fault Diagnosis for DSL-Based IPTV Networks using the Renyi entropy Proc. IEEE Globecom'2012, Dec. 3-7, Anaheim, USA




  • H. Fanaee, M. Oliveira, J. Gama, S. Malinowski and R. Morla Event and anomaly detection using Tucker3 decomposition Proc. UDM'2012 Workshop, Aug. 27-31, Montpellier, France
  • pdf-icon S. Malinowski, X. Artigas, C. Guillemot and L. Torres Distributed coding using punctured quasi-arithmetic codes for memory and memoryless sources Proc. IEEE PCS'2009, May 6-8, Chicago, Illinois, USA
  • pdf-icon X. Artigas, S. Malinowski, C. Guillemot and L. Torres Overlapped arithmetic codes with memory Proc. Eusipco'2008, August 25-29, Lausanne, Switzerland
  • pdf-icon X. Artigas, S. Malinowski, C. Guillemot and L. Torres Overlapped quasi-arithmetic codes for distributed video coding Proc. IEEE ICIP'2007, Sept. 16-19, San Antonio, Texas, USA
  • pdf-icon S. Malinowski, H. Jégou and C. Guillemot Error recovery properties of quasi-arithmetic codes and soft decoding with length constraint Proc. IEEE ISIT'2006, July 9-14, Seattle, Washington, USA
  • pdf-icon S. Malinowski, H. Jégou and C. Guillemot On the link between the synchronization recovery and soft decoding of Variable Length Codes Proc. ITG'2006, April 3-7, Munich, Germany
  • pdf-icon H. Jégou, S. Malinowski and C. Guillemot Trellis state aggregation for soft decoding of variable length codes Proc. IEEE SIPS'2005, November 2-4, Athens, Greece. Best student paper award. Source code(4kB) or Linux exe(400kB,i386-P4)

National conferences





  • H. Jégou, S. Malinowski and C. Guillemot Décodage conjoint source/canal par agrégation d'états et décodage multi-treillis Proc. CORESA'2005, November 6-7, Rennes, France

Master Thesis




  • S. Malinowski Décodage souple de Codes à Longueur Variable Master Thesis in the TEMICS project (IRISA)

Teaching (french version only)

AgroCampus Ouest, Rennes (2012-2013)

  • Algorithmique et Programmation en Python - 33h eq. TD, dont 8h CM
    Niveau: 2ème année diplôme Ingénieur Agronome, équivalent M1
    Mots clés: bases de l'algorithmique, utilisation de Python

  • Algorithmique et Programmation en Python, application au traitement d'images - 40h eq. TD, dont 12h CM
    Niveau: Master 1 Géographie, Parcours Image
    Mots clés: bases de l'algorithmique, utilisation de Python, traitement d'images

  • Bases de données et programmation en R - 50h eq. TD
    Niveau: 1ère année diplôme Ingénieur Agronome, équivalent L3
    Mots clés: bases de données relationelles, SQL, algorthmique, utilisation de R

  • Apprentissage automatique - 15h eq. TD, dont 6h CM
    Niveau: Master 2 Géographie, Parcours Image
    Mots clés: méthodes d'apprentissages supervisées et non supervisées, application images de télédétection

  • Méthodes de développement d'applications informatiques, 16h eq.TD
    Niveau: 2ème année diplôme Ingénieur Agronome, équivalent M1 Mots clés: concepts objets, diagrammes d'UML

  • Programmation objet avec Java, 33h eq.TD, dont 8h CM
    Niveau: 2ème année diplôme Ingénieur Agronome, équivalent M1 Mots clés: modélisation objet, héritage, polymorphisme, langagae Java

Univ. Rennes 1 et classes préparatoires scientifiques, lycée Chateaubriand, Rennes (2005 - 2010)

  • Compression d'images et vidéos - 45h eq.TD dont 24h CM
    Niveau: 3ème année diplôme Ingénieur ESIR Rennes (eq. Master 2)
    Mots clés: Théorie de l'information, codage entropique, normes JPEG-MPEG

  • Traitement d'images (16h eq.TD)
    Niveau: 2ème année diplôme Ingénieur ESIR Rennes (eq. Master 1)
    Mots clés: filtrage, transformée de Fourier, ondelettes, morphologie, segmentation, seam carving

  • Vision (25h eq.TD)
    Niveau: 3ème année diplôme Ingénieur ESIR Rennes (eq. Master 2)
    Mots clés: chaînes de Markov cachées, estimation, descripteurs d'image...

  • Introduction à l'algorithmique et programmation(~200h eq.TD)
    Licence 1, Licence 2 et Master CCI, classes préparatoires scientifique
    Mots clés: variables, structures des programmes, boucles, structure de données, Java, C, Mathematica, Scilab

  • Bureautique (32h eq.TD)
    Licence 1
    Mots clés: traitement de texte, tableur, diaporamas

  • Bases de données (10h eq.TD)
    Master CCI
    Mots clés: algèbre relationnelle, requêtes, SQL, partage de données, concurrence d'accés

  • Statistiques (6h eq.TD)
    1ère année diplôme Ingénieur ESIR Rennes (eq. Licence 3)
    Mots clés: statistiques descriptives, lois usuelles, corrélation, test de corrélation, test d'hypothèses

Flag Counter