Canada.ca
In 1996 CIC published the Canadian Language Benchmarks: English as a Second The second version
Pluriverse: A Post-Development Dictionary
23 ago 2012 Cover illustration: Ashish Kothari. Printed at Chaman Offset Delhi 110 002. PLURIVERSE Prelims (pp. i-x).indd 4. 4/12/2019 12:24:11 PM ...
Product Data: PULSE Sound Quality Software Type 7698 (bp1589)
benefit of OLE automation the software can be controlled from other Jury Test tool for designing and executing sound quality listening tests.
A STUDY OF FULVIA
Fulvia. Bengtson's biography of Antony Marcus Antonius
Transfer Learning methods for temporal data
1 feb 2022 tous les membres du jury pour avoir accepté de participer `a ma soutenance de ... 4 Domain adaptation with multiple sources in regression.
How to set up a Workers Co-op
Third edition written by Footprint Workers' Co-operative Ltd and Seeds for Change 4. Getting your co-op together. 4.1 Finding the right people.
DAVID MORGAN. LEXICON LATINVM
haedînum
handle with kid gloves molli bracchio tractare [Cic.] dintendant de secrétaire du palais impérial
" Benoît; ...
CATALOG
23 ago 2017 4. Catalog 2017. 2018. Office of Research and Graduate Studies ... Bou Saber Maan
Armor July-August 1985 Edition
the postmaster. USPS 467-970. JUly-AUgUSt lge5 VOl XClV NO. 4. FEATURES for a new Armored Maintenance Vehicle (AMV) de-.
The Civil Engineering Handbook Second Edition
The second edition of the Civil Engineering Handbook has been revised and updated to Construction (CIC) • Toward Advanced Construction Automation •.
Thèse de doctorat
NNT: 2021UPASM037Methodes d'apprentissage statistique de type "Transfer Learning" pour des donnees temporelles multivarieesTransfer Learning for Temporal Data
These de doctorat de l'Universite Paris-Saclay
Ecole doctorale n574, Ecole Doctorale de Mathematiques Hadamard (EDMH)Specialite de doctorat: Mathematiques Appliquees
Unite de recherche: Centre Borelli
Referent: ENS Paris-Saclay
These presentee et soutenue a Gif-sur-Yvette,
le 7 octobre 2021, parGuillaume RICHARDComposition du jury
Younes BennaniPresident et RapporteurProfesseur, Universite Sorbonne Paris Nord (Laboratoired'Informatique de Paris NordGianluca BontempiRapporteurProfesseur, Universite Libre de Bruxelles (Departement
d'Informatique)Jairo CugliariExaminateurMaitre de Confeerences, Universite Lumiere 2 (Laboratoire ERIC)
Michele SebahExaminatriceDirectrice de Recherche, INRIA (Equipe TAO)Direction de la these
Mathilde MougeotDirectriceProfesseur, ENS Paris-Saclay (Centre Borelli) et ENSIIE Nicolas VayatisDirecteurProfesseur, ENS Paris-Saclay (Centre Borelli) Georges HebrailTuteur en entrepriseChercheur Senior, EDF R&DRemerciements
Mes premiers remerciements vont
`a mes directeurs de th`ese, Mathilde Mougeot et Nicolas Vayatis pour leur con- fiance au long de ces trois ann ´ees de th`ese, et de m"avoir donn´e l"opportunit´e de travailler sur des sujets aussi int´eressants. Je remercie´egalement Georges H´ebrail qui a su m"accompagner d`es mes premiers pas chez EDF
et d´epasser son rˆole d"encadrant industriel pour toujours me donner des bons conseils. Je remercie´egalement
tous les membres du jury pour avoir accept ´e de participer`a ma soutenance de th`ese. En particulier, je remercie Youn`es Bennani qui a accept´e de pr´esider le jury en plus d"avoir rapport´e ma th`ese, Gianluca Bontempi qui a lui
aussi rapport ´e ma th`ese et enfin Jairo Cugliari et Mich`ele Sebag en tant qu"examinateurs.Ensuite, je souhaite remercier toute l"
´equipe SOAD d"EDF pour tous les bons moments pass´es ensemble. J"aurai certainement l"occasion de le dire de vive voix, mais c" ´etait un r´eel plaisir de travailler avec vous. En plus de Georges, j"aimerais remercier Ghislain qui m"a accompagn ´e sur la fin de la th`ese. Quoiqu"il en soit, j"esp`ere qu"on sera amen ´e`a retravailler ensemble et que je serai toujours invit´e au Diamant! Je veux aussi remercier l"ensemble des membres du Centre Borelli, en particulier les th´esards. L"entraide entre
les doctorants et nos discussions ont permis d" ´egayer les moments difficiles. En particulier, je remercie Antoine avec qui c" ´etait un plaisir de collaborer et je te souhaite le meilleur pour le reste de ta th`ese. Enfin, je veux remercier mes amis et ma famille, qui m"ont soutenu tout au long de la th `ese et m"ont aid´e`a me changer les id´ees. En particulier, je remercie ma m`ere, mon p`ere et mon petit fr`ere qui m"ont fait grandir, m"ont d´ej`a
accompagn´e dans toutes les´etapes importantes de ma vie et continueront`a le faire. Finally, I want to thank you,
Qu`ynh Anh, for your love and support. It really made it easier to overcome every difficult moment and to move to
the next chapter together. iContents
R´esum´e (en franc¸ais)11
1 Introduction19
1.1 Motivation
191.2 Organization of the manuscript
212 Background on Transfer Learning
252.1 What is Transfer Learning?
262.2 Theory of Domain Adaptation
282.2.1 Generalization bounds
292.2.2 Divergence-based Domain Adaptation
312.2.3 Alternative approaches
362.2.4 Summary
382.3 Existing approaches for Homogeneous Transfer Learning
382.3.1 Instance-based domain adaptation
392.3.2 Feature-based domain adaptation
422.3.3 Alternative methods
442.3.4 Summary
452.4 Learning from Multiple Sources
462.4.1 Multi-Task Learning and Domain Generalization
472.4.2 Multi-source domain adaptation
492.5 Conclusion
513 Deep Time Series Representations for Non-Intrusive Load Monitoring
533.1 Background on Time Series Representations
553.1.1 Framework
553.1.2 Overview of Univariate Time Series Representations
551
3.2 Transferability of Deep Time Series Representations. . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
3.2.1 Deep Time Series Representations
583.2.2 On Transferability of Deep Time Series Representations
613.3 Transfer Learning in Non Intrusive Load Monitoring
613.3.1 General presentation
613.3.2 Review of methods
633.3.3 Datasets
643.3.4 Problem formulation
663.4 Time Series Normalization for Invariant Appliance Recognition
683.4.1 Global and z-normalization
683.4.2 Normalization for appliance consumption
703.4.3 Model
723.5 Experiments on NILM Datasets
733.5.1 Preprocessing and Methods
743.5.2 Same House
773.5.3 Cross-House Results
773.5.4 Cross-Dataset Results
793.5.5 Discussion
813.6 Conclusion
824 Domain adaptation with multiple sources in regression
854.1 Domain Adversarial Learning withH-divergence. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
4.1.1 Literature review
874.1.2 Limits of Domain Adversarial Adaptation in Regression withH-divergence. . . . . . . . . . . . 88
4.2 Hypothesis-Discrepancy for Domain Adaptation in Regression
894.2.1 Hypothesis-Discrepancy
894.2.2 Domain Adaptation Guarantees with Hypothesis-Discrepancy
914.3 Minimizing the hypothesis-discrepancy
924.4 Extension to multiple sources
944.4.1 Theoretical Guarantees with multiple sources
944.4.2 Algorithm
974.5 Experiments
984.5.1 Synthetic data
994.5.2 Appliance Consumption Estimation
1022
4.5.3 Same-house results. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 104
4.5.4 Cross-house results
1054.5.5 Experiments on other datasets
1084.6 Extension to semi-supervised adaptation
1114.7 Conclusion
1145 Covariance-based Transfer Learning with applications to Multivariate Time Series
1155.1 Outline of the method
1165.2 Multivariate Time Series and Covariance
1185.3 Riemannian Geometry of Symmetric Positive Definite Matrices and Time Series
1205.3.1 Basics
1205.3.2 Working with time series
1225.3.3 Statistical Learning with SPD Matrices
1235.4 Transferable subspace using Covariance information
1235.4.1 Framework
1235.4.2 Learning a subspace aligning domains
1245.4.3 Related works
1275.4.4 Algorithm
1285.4.5 Hyperparameter Selection
1305.5 Numerical Results
1325.5.1 Simulated data
1325.5.2 Human Activity Recognition
1355.6 Conclusion
1416 Conclusion and Perspectives143
A Neural Networks147
B Clustering consumer consumption with auto-encoders 151B.1 Data presentation
151B.2 Method
152B.2.1 Convolutional AutoEncoder
152B.2.2 Compared methods
153B.2.3 Outliers
154B.3 Results
1553
C Implementations159
C.1 Public implementations
159C.2 Other implementations
160C.2.1 List of statistical features extracted (Chapter 3 160
C.3 Additional Experiments of Chapter
3 162C.4 Details about implementations of Chapter
4 1634
List of Figures
1 Illustration du NILM:
`a partir de la consommation totale de la maison, l"objectif est de retrouver la consommation de chaque appareil. 131.1 Non Intrusive Load Monitoring illustration: from the whole house consumption, the goal is to retrieve
the consumption of each appliance 202.1 Taxonomy of Transfer Learning
272.2 Illustration of theH-divergence with linear classifiers. . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.3 Illustration of the discrepancy with linear regressorsH=fh:x!wTx;kwk21gand`2-loss.
Source and target data are generated as 1D-Gaussian distributions centered on1and+1withdifferent standard deviations in each graph (f1;1g,f0:5;5gandf4;4g).. . . . . . . . . . . . . . . . . . 36
2.4 Domain Adversarial Neural Network (figure from [
7244
2.5 Scenarios of Transfer Learning with or without multiple sources
513.1 Euclidean Distance vs Dynamic Time Warping
563.2 Different architectures for time series classification
593.3 Convolutional Auto-Encoder
603.4 Sub-problems of Non Intrusive Load Monitoring (NILM)
633.5 Examples of monitored appliances for each house: a blue square means the appliances was moni-
tored. A white square means it was not monitored or was mixed with another appliance. 653.6 Examples of monitored consumption for one day in the Electric DataBase (x-axis is in hours and
y-axis is in W) 663.7 Examples of monitored consumption for one day in the REFIT Database (x-axis is in hours and y-axis
is in W ; the house numbers correspond to the ones in the database as technical issues happened for other houses) 673.8 Examples of signatures from the Trace Base dataset (x-axis is in minutes, y-axis is in W). Signatures
have been zero-padded to a length of 2 hours. 69quotesdbs_dbs27.pdfusesText_33
[PDF] Biographies des membres du jury fédéral de la danse Présidente
[PDF] Biographies du Parc et du Château de Champs-sur
[PDF] Biographies EUROPE.rtf
[PDF] Biographies Felipe Sauvageon: Rumbas - i - Mexique Et Amérique Centrale
[PDF] Biographies individuelles
[PDF] biographies of our new partners/ associate partners/ directors.
[PDF] Biographies of speakers and moderators NAME - MGIEP - France
[PDF] biographies of the candidates
[PDF] Biographies: Policy Dialogue with Civil Society
[PDF] BIOGRAPHIE_files/CV site
[PDF] Biographie_files/Liste de mes expos - Anciens Et Réunions
[PDF] Biographie– Prof. Dr. Emmanuelle Charpentier - Max-Planck
[PDF] Biography -
[PDF] biography - Alison Jacques Gallery