I was
appointed Full Professor in 2004. I obtained my
accreditation to lead research (HDR) in 2002 and a PhD
in Computer Science in 1998.
I completed my "Classes Préparatoires" (Competitive entrance to French
Engineering Schools) in 1988-90, after which
I got my "Ingénieur Agro" degree (MSc in Agronomical Engineering
Sciences, majoring in industrial microbiology and biostatistics)
and a MSc in Computer Science (both magna cum laude) in 1993.
Over the past ten years I have spent approximately 10% of my time as
an invited researcher of other labs, including Sony Computer Science
Laboratories, Inc. (Tokyo, five times), Helsinki U., Ottawa U.,
LIX-Ecole Polytechnique (Palaiseau),
LaHC-U. Saint-Etienne, I3S-U. Nice.

I received in 2013 the "Grand
prix ANR du numérique" --- Press
release (ANR Digital Technology Award; ANR
= French NSF).

Portfolio allocation

Portfolio allocation theory has been heavily influenced by the mean-variance approach of Harry Markowitz. With colleagues from finance, maths and economics, we have alleviated the Gaussian assumption of the model, derived the exact expression of the risk premium --- which turns out to be a Bregman divergence --- and certainty equivalent in this generalized model, and finally devised in this broader exact model an on-line learning algorithm with guaranteed lowerbounds on its cumulated certainty equivalents.

ICML 2011 |

Distribution of incomes and information
geometry

Inequality indices evaluate the divergence between the income distribution and the hypothetical situation where all individuals receive the mean income, and are unambiguously reduced by a Pigou-Dalton progressive transfer. With a colleague from economics, we have characterized the unique class of divergence measures between income distributions that is consistent with popular views of normative economics. It appears to match Bregman divergences (to appear, the Journal of Economic Theory).

JET 2011 |

Computational information geometry

With colleagues of maths and computer science, we have studied topological spaces associated with distortion functions that are not metrics: Bregman divergence --- thus that do not meet neither symmetry nor the triangular inequality in the general case ---. Such non-metric spaces are fundamental in statistics, information geometry, classification, etc. . We have considered a large number of problems and algorithms previously known for metric spaces, that we have lifted to these spaces.

DCG 2010 | TIT 2009 | ECML 2008 | SODA 2007 |

Boosting

A very active supervised learning trend has been flourishing over the last decade: it studies functions known as surrogates --- upperbounds of the empirical risk, generally with particular convexity properties ---, whose minimization remarkably impacts on empirical / true risks minimization. Surrogates play fundamental roles in some of the most successful supervised learning algorithms, including AdaBoost, additive logistic regression, decision tree induction, Support Vector Machines. Our contributions include new boosting algorithms, unification of popular boosting algorithms, formal convergence bounds.

IJCV 2012 | TPAMI 2009 | NIPS 2008 | IJCAI 2007 | AIJ 2007 | ICML 2004 |

Image segmentation

The field of image segmentation is a very active field which seeks to make a partition of an image into regions a user would consider as perceptually distinct. We have created a particular blend of statistics and algorithmics whose error is formally limited from both the qualitative and quantitative standpoints. The algorithm --- which is approximately time and space linear --- can be implemented in straightforward ways, and can be extended to numerous situations where e.g. a user bias is available, images are highly corrupted or occluded, etc. . This algorithm, known as SRM for "Statistical Region Merging", is being used in quite a large number of applications and fields.

Sources | PRJ 2005 | MM 2005 | TPAMI 2004 | CVPR 2004 | CVPR 2003 | CVPR 2001 |

The structural complexity of learning

The most popular models of learning, like the PAC model of Valiant, assume time complexity constraint over the algorithms. The lack of known algorithms for particular classifier has naturally questioned whether such algorithms really do exist. Using original reductions known as self-improving --- because they are made from a problem onto itself while blowing-up its hardness --- we have proven numerous inapproximability results for various classifiers, some even translating into negative weak learning results. We kept during a decade or so the largest inapproximability ratio for learning DNF (our ISAAC paper), a popular problem in the late XXth century.

TCS 2007 | TCS 2003 | ALT 2000 | ALT 1999 | ISAAC 1998 | ILP 1998 | ICML 1996 |

Email: .

Phone: (+596)
596 72 74 04 (Caribbean area --- UTC - 4h)

I have published 100+ papers in areas covering Classification/Learning, Computational Complexity, Economics, Finance, Image processing, Information Geometry, Statistics and Probability.

References include IJCAI/UAI/ECAI, NIPS/ALT, ICML/ECML, SODA, CVPR (for conferences), and AIJ/JAIR, JMLR, TPAMI, TIT, IJCV, the Journal of Economic Theory, the Journal of Physics (for journals). A rather incomplete list of publications is provided below. You may also be interested by my Google Scholar records, or my DBLP publication list.

Feel free to contact me for questions or remarks about my works, to obtain reprints or source codes.

- Natalia Polouliakh,
*Richard Nock*, Frank Nielsen and Hiroaki Kitano*Gene Clustering Program, Gene Clustering Method, and Gene Cluster Analyzing Device*

Labs: Sony Computer Science Laboratories, Inc. (NP, FN, HK), CEREGMIA --- UAG (RN).*Serial number*: JP2010157214, US2011246080, EP2354988, CN102227731, 2010-11

#### [ Journals ] ( Patents: above ) ( Conferences: below )

*Richard Nock*, Paolo Piro, Frank Nielsen, Wafa Bel Haj Ali and Michel Barlaud*Boosting k-NN for Categorization of Natural Scenes***International Journal of Computer Vision**

(100), pp 294-314, 2012.*Springer*.

Related material (paper): here.

- Frank Nielsen and
*Richard Nock**A closed-form Expression for the Sharma-Mittal Entropy of Exponential Families***Journal of Physics A: Mathematical and Theoretical**

(45)-3, 9 pp, 2011.*IOP*.

Related material (paper): here.

- Brice Magdalou and
*Richard Nock**Income Distributions and Decomposable Divergence Measures***Journal of Economic Theory**

(146)-6, pp 2440-2454, 2011.*Elsevier*.

Related material (paper): here.**Top 25 hottest JET papers, 2011 (more)**.

- Frank Nielsen and
*Richard Nock**Skew Jensen-Bregman Voronoi Diagrams***Transactions on Computational Science**

(14), pp 102-128, 2011.*Springer*.

Related material (paper): here.

- Paolo Piro,
*Richard Nock*, Frank Nielsen and Michel Barlaud*Leveraging k-NN for Generic Classification Boosting***Neurocomputing**

(80), pp 3-9, 2012.*Elsevier*.

Related material (paper): here.

- Jean-Daniel Boissonnat, Frank Nielsen and
*Richard Nock**Bregman Voronoi Diagrams***Discrete and Computational Geometry**

(44)-2, pp 281-307, 2010.*Springer Verlag*.

Related material (paper): here.

- Frank Nielsen and
*Richard Nock**Sided and Symmetrized Bregman Centroids***IEEE Transactions on Information Theory**

(55)-6, pp 2882-2904, 2009.*IEEE CS Press*.

Related material (paper): here.

*Richard Nock*and Frank Nielsen*Bregman Divergences and Surrogates for Learning***IEEE Transactions on Pattern Analysis and Machine Intelligence**

(31)-11, pp 2048-2059, 2009.*IEEE CS Press*.

Related material (paper): here.

- Natalia Polouliakh,
*Richard Nock*, Frank Nielsen and Hiroaki Kitano*G-Protein Coupled Receptor Signaling Architecture of Mammalian Immune Cells***PLoS ONE**

(4)-1, e4189, 2009.*Public Library of Sciences*.

Related material (paper): here.

*Richard Nock*, Pascal Vaillant, Claudia Henry and Frank Nielsen*Soft Memberships for Spectral Clustering, with Application to Permeable Language Distinction***Pattern Recognition**

(42)-1, pp 43-53, 2009.*Elsevier*.

Related material (paper): here.

- Frank Nielsen and
*Richard Nock**Approximating Smallest Enclosing Balls with Application to Machine Learning***International Journal on Computational Geometry and Applications**

(19)-4, pp 389-414, 2009.*World Scientific Publishing*.

Related material (paper): here.

- Frank Nielsen and
*Richard Nock**On the Smallest Enclosing Information Disk***Information Processing Letters**

(105)-3, pp 93-97, 2008.*Elsevier*.

Related material (paper): here.

*Richard Nock*and Frank Nielsen*Self-Improved gaps Almost Everywhere for the Agnostic Approximation of Monomials***Theoretical Computer Science (A)**

(377)-1-3, pp 139-150, 2007.*Elsevier*.

Related material (paper): here.

*Richard Nock*and Frank Nielsen*A Real Generalization of discrete AdaBoost***Artificial Intelligence**

(171)-1, pp 25-41, 2007.*Elsevier*.

Related material (paper): here.

- Pierre-Alain Laur,
*Richard Nock*, Jean-Emile Symphor and Pascal Poncelet*Mining Evolving Data Streams for Frequent Patterns***Pattern Recognition**

(40)-2, pp 492-503, 2007.*Elsevier*.

Related material (paper): here.

- Pierre-Alain Laur, Jean-Emile Symphor,
*Richard Nock*and Pascal Poncelet*Statistical Supports for Mining Sequential Patterns and Improving the Incremental Update Process on Data Streams***International Journal on Intelligent Data Analysis**

(11)-1, pp 29-47, 2007.*IOS press*.

Related material (paper): here.

*Richard Nock*and Frank Nielsen*On Weighting Clustering***IEEE Transactions on Pattern Analysis and Machine Intelligence**

(28)-8, pp 1223-1235, 2006.*IEEE CS press*.

Related material (paper): here.

Related material (page with sources, binaries, images, etc.): here.

*Richard Nock*and Frank Nielsen*Semi-supervised Statistical Region Refinement for Color Image Segmentation***Pattern Recognition**

(38)-6, pp 835-846, 2005.*Elsevier*.

Related material (paper): here.

Related material (page with sources, images, etc.): here.

- Frank Nielsen and
*Richard Nock**A Fast Deterministic Smallest Enclosing Disk Approximation Algorithm***Information Processing Letters**

(93)-6, pp 263-268, 2005.*Elsevier*.

Related material (paper): here.

*Richard Nock*and Frank Nielsen*Statistical Region Merging***IEEE Transactions on Pattern Analysis and Machine Intelligence**

(26)-11, pp 1452-1458, 2004.*IEEE CS press*.

Related material (paper): here.

Related material (page with sources, images, etc.): here.

*Richard Nock*and Frank Nielsen*On Domain-Partitioning Induction Criteria: Worst-case Bounds for the Worst-case Based***Theoretical Computer Science (A)**

(321), pp 371-382, 2004.*Elsevier*.

Related material (paper): here.

*Richard Nock**Complexity in the Case against Accuracy Estimation***Theoretical Computer Science (A)**

(301), pp 143-165, 2003.*Elsevier*.

Related material (paper): here.

*Richard Nock*, Marc Sebban and Didier Bernard*A Simple locally Adaptive Nearest Neighbor Rule with Application to Pollution Forecasting***International Journal on Pattern Recognition and Artificial Intelligence**

(17)-8, pp 1-14, 2003.*World Scientific Publishing*.

Related material (paper): here.

*Richard Nock*, Tapio Elomaa and Matti Kääriäinen*Reduced Error Pruning of Branching Programs cannot be Approximated to within a Logarithmic Factor***Information Processing Letters**

(87), pp 73-78, 2003.*Elsevier*.

Related material (paper): here.

- Marc Sebban,
*Richard Nock*and Stéphane Lallich*Stopping Criterion for Boosting-based Data Reduction Techniques: from Binary to Multiclass Problems***Journal of Machine Learning Research**

(3), pp 863-885, 2002.*MIT Press*.

Related material (paper): here.

*Richard Nock**Inducing Interpretable Voting Classifiers without trading Accuracy for Simplicity: Theoretical Results, Approximation Algorithms, and Experiments***Journal of Artificial Intelligence Research**

(17), pp 137-170, 2002.*Morgan Kauffman*.

Related material (paper): here.

- Marc Sebban and
*Richard Nock**A Hybrid Filter/Wrapper Approach of Feature Selection Using Information Theory***Pattern Recognition**

(35) 4, pp 835-846, 2002.*Elsevier*.

Related material (paper): here.

*Richard Nock*and Marc Sebban*Advances in Adaptive Prototype Weighting and Selection***Artificial Intelligence Tools**

(10), pp 137-156, 2001.*World Scientific Pub.*.

- Marc Sebban,
*Richard Nock*, Jean-Hugues Chauchat and Ricco Rakotomalala*Impact of Learning Set Quality and Size on Decision Tree Performances***International Journal of Computers, Systems and Signals**

pp 85-105, 2001.*IAAMSAD Publishing*.

*Richard Nock*and Marc Sebban*An improved bound on the Finite Sample Risk of the Nearest Neighbor Rule***Pattern Recognition Letters**

(22) 3-4. pp 413-419, 2001.*Elsevier*.

Related material (paper): here.

*Richard Nock*and Marc Sebban*A Bayesian Boosting Theorem***Pattern Recognition Letters**

(22) 3-4. pp 407-412, 2001.*Elsevier*.

Related material (paper): here.

*Richard Nock*and Pascal Jappy*Decision Tree based induction of Decision Lists***International Journal on Intelligent Data Analysis**

(3) 3. pp 227-240, 1999.*Elsevier-IOS Press*.

- Olivier Gascuel,
*Richard Nock**et al.*(SYMENU Group)*Twelve Numerical, Symbolic and Hybrid Supervised Classification Methods***International Journal of Pattern Recognition and Artificial Intelligence**

pp 517-572, 1998.*World Scientific Publishing*.

#### Book Chapters

*Richard Nock*, Brice Magdalou, Eric Briys and Frank Nielsen*Mining Matrix Data with Bregman Matrix Divergences for Portfolio Selection*

Chapter of the book "**Matrix Information Geometry**" (*Ed. F. Nielsen and R. Bathia*)*accepted*, 2012.*Springer*.

Related material (paper): here.

- Paolo Piro, Michel Barlaud,
*Richard Nock*and Frank Nielsen*k-NN boosting prototype learning for object classification*

Chapter of the book "**Analysis, Retrieval and Delivery of Multimedia Contents**" (*Ed. N. Adami, A. Cavallaro, R. Leonardi and P. Migliorati*)*accepted*, 2012.*Springer*.

- Eric Briys, Brice Magdalou and
*Richard Nock**Portfolios, Information and Geometry: Simplex Orbis non Sufficit*

Chapter of the book "**After the Crisis: Rethinking Finance**" (*Ed. T. Lagoarde-Segot*)

pp 225-244, 2010.*Nova publishers*.

*Richard Nock*and Marc Sebban*Prototype Selection using Boosted Nearest-Neighbors*

Chapter of the book "**Instance Selection and Construction for Data Mining**" (*Ed. H. Liu and H Motoda*)

pp 301-318, 2001.*Kluwer Academic Publishers*.

#### [ Conferences ] ( Patents: above ) ( Journals: above )

*Richard Nock*, Frank Nielsen and Eric Briys*Non-linear Book Manifolds : learning from Associations the Dynamic Geometry of Digital Libraries***JCDL'13 - ACM/IEEE International Joint Conferences on Digital Libraries**(*Indianapolis, USA*)*accepted (long paper)*, 2013.*ACM Press*.

Related material (paper): here.

*Richard Nock*and Frank Nielsen*Information-Geometric Lenses for Multiple Foci+Contexts Interfaces***SIGGRAPH Asia'13 - ACM SIGGRAPH Conference on Computer Graphics and Interactive Techniques in Asia**(*Hong Kong, PRC*)*accepted (technical brief)*, 2013.*ACM Press*.

Related material (paper): here.

- Roberto D'Ambrosio,
*Richard Nock*, Wafa Bel Haj Ali, Frank Nielsen and Michel Barlaud*Boosting nearest neighbors for the efficient estimation of posteriors***ECML'12 - European Conference on Machine Learning**(*Bristol, UK*)

pp 314--329, 2012.*Springer Verlag LNCS 7523*.

- Wafa Bel Haj Ali, Paolo Piro, Dario Giampaglia, Thierry Pourcher,
*Richard Nock*and Michel Barlaud*Classification of Biological Cells using bio-inspired descriptors***ICPR'12 - International Conference on Pattern Recognition**(*Tsukuba, Japan*)

pp 3353-3357, 2012.*IAPR / IEEE press*.

*Richard Nock*, Brice Magdalou, Eric Briys and Frank Nielsen*On Tracking Portfolios with Certainty Equivalents on a Generalization of Markowitz Model: the Fool, the Wise and the Adaptive***ICML'11 - International Conference on Machine Learning**(*Seattle, Washington*)

pp 73-80, 2011.*Omnipress*.

Related material (paper): here.

- Paolo Piro,
*Richard Nock*, Frank Nielsen and Michel Barlaud*Multi-class Leveraged k-NN for Image Classification***ACCV'10 - Asian Conference on Computer Vision**(*Queenstown, New Zealand*)

pp 67-81, 2010.*Springer Verlag LNCS 6494*.

- Vincent Garcia, Frank Nielsen and
*Richard Nock**Hierarchical Gaussian Mixture Model***ICASSP'10 - International Conference on Acoustics, Speech and Signal Processing**(*Dallas, Texas*)

pp 4070-4073, 2010.*IEEE SP Press*.

- Frank Nielsen and
*Richard Nock**Entropies and cross-entropies of exponential families***ICIP'10 - International Conference on Image Processing**(*Hong Kong, China*)

pp 3621 - 3624, 2010.*IEEE SP press*.

- Paolo Piro,
*Richard Nock*, Frank Nielsen and Michel Barlaud*Boosting Bayesian MAP Classification***ICPR'10 - International Conference on Pattern Recognition**(*Istambul, Turkey*)

pp 661-665, 2010.*IAPR / IEEE press*.

- Frank Nielsen and
*Richard Nock**Jensen-Bregman Voronoi Diagrams and Centroidal Tessellations***ISVD'10 - International Symposium on Voronoi Diagrams**(*Quebec, Canada*)

pp 56-65, 2010.*IEEE CS Press*.

- Frank Nielsen and
*Richard Nock**The Dual Voronoi Diagrams with Respect to Representational Bregman Divergences***ISVD'09 - International Symposium on Voronoi Diagrams**(*Copenhagen, Denmark*)

pp 71-78, 2009.*IEEE CS Press*.

- Vincent Garcia, Frank Nielsen and
*Richard Nock**Levels of details for Gaussian mixture models***ACCV'09 - Asian Conference on Computer Vision**(*Xi'ian, China*)

pp 514-525, 2009.*Springer Verlag LNCS 5995*.

*Richard Nock*and Frank Nielsen*Intrinsic Geometries in Learning***ETVC'08 - Emerging Trends in Visual Computing**(*Ecole Polytechnique*, 11/18-20/08)

pp 175-215, 2008.*Springer Verlag LNCS 5416*.

Related material (paper): here.**Invited paper**.

- Frank Nielsen and
*Richard Nock**Clustering Multivariate Normal Distributions***ETVC'08 - Emerging Trends in Visual Computing**(*Ecole Polytechnique*, 11/18-20/08)

pp 164-174, 2008.*Springer Verlag LNCS 5416*.**Invited paper**.

*Richard Nock*and Frank Nielsen*On the Efficient Minimization of Classification-Calibrated Surrogates***NIPS*21 - Advances in Neural Information Processing Systems**(*Vancouver, Canada*)

pp 1201-1208, 2008.*MIT Press*.

Related material (paper): here.**Spotlight paper**.

*Richard Nock*, Panu Luosto and Jyrki Kivinen*Mixed Bregman Clustering with Approximation Guarantees***ECML'08 - European Conference on Machine Learning**(*Antwerp, Belgium*)

pp 154-169, 2008.*Springer Verlag LNCS*.

- Frank Nielsen and
*Richard Nock**Quantum Voronoi Diagrams and Holevo Channel Capacity for 1-Qubit Quantum States***ISIT'08 - IEEE International Symposium on Information Theory**(*Toronto, Canada*)

pp 96-100, 2008.*IEEE CS Press*.

- Natalia Polouliakh, Yukiko Matsuoka, Samik Ghosh,
*Richard Nock*, Frank Nielsen, Satoshi Kitajima, Atsuya Takagi, Ken-Ichi Aisaki, Jun Kanno and Hiroaki Kitano*Signaling Network in Mouse Embryonic Stem Cells***ISCB'08 - International Conference on Systems Biology**(*Göteborg, Sweden*)

2008 (poster).

- Frank Nielsen and
*Richard Nock**Bregman sided and symmetrized centroids***ICPR'08 - International Conference on Pattern Recognition**(*Tampa, USA*)

pp 1-4, 2008.*IAPR / IEEE press*.

*Richard Nock*and Frank Nielsen*On the Efficient Minimization of Convex Surrogates in Supervised Learning***ICPR'08 - International Conference on Pattern Recognition**(*Tampa, USA*)

pp 1-4, 2008.*IAPR / IEEE press*.**Best Scientific Paper Award**.

- Claudia Henry,
*Richard Nock*and Frank Nielsen*Real Boosting*à la carte*with an application to Boosting Oblique Decision Trees***IJCAI'07 - International Joint Conference on Artificial Intelligence**(*Hyderabad, India*)

pp 842-847 (oral), 2007.*Morgan Kaufmann*.

Related material (paper): here.

- Frank Nielsen, Jean-Daniel Boissonnat and
*Richard Nock**On Bregman Voronoi Diagrams***SODA'07 - ACM/SIAM International Symposium on Discrete Algorithms**(*New Orleans, USA*)

pp 746-755, 2007.*ACM press*.

- Frank Nielsen, Jean-Daniel Boissonnat and
*Richard Nock**Visualizing Bregman Voronoi Diagrams***SoCG'07 - ACM International Symposium on Computational Geometry**(*Gyeongju, Korea*)

pp 121-122, 2007.*ACM press*.

- Frank Nielsen and
*Richard Nock**Fast Graph Segmentation based on Statistical Aggregation Phenomena***MVA'07 - IAPR International Conference on Machine-Vision Applications**(*Tokyo, Japan*)

pp 150-153, 2007.*IAPR press*.

*Richard Nock*and Frank Nielsen*A Real Generalization of discrete AdaBoost***ECAI'06 - European Conference on Artificial Intelligence**(*Riva del Gardia, Italy*)

pp 509-515, 2006.*IOS press*.**Best Paper Award**.

*Richard Nock*, Pascal Vaillant, Frank Nielsen and Claudia Henry*Soft Uncoupling of Markov chains for Permeable Language Distinction: A New Algorithm***ECAI'06 - European Conference on Artificial Intelligence**(*Riva del Gardia, Italy*)

pp 823-824, 2006.*IOS press*.

- Frank Nielsen and
*Richard Nock**On Approximating the Smallest Enclosing Bregman Balls***SoCG'06 - ACM International Symposium on Computational Geometry**(*Sedona, USA*)

pp 485-486, 2006.*ACM press*.

*Richard Nock*, Pierre-Alain Laur and Jean-Emile Symphor*Statistical Borders for Incremental Mining***ICPR'06 - International Conference on Pattern Recognition**(*Hong-Kong, China*)

pp 212-215, 2006.*IAPR / IEEE press*.

- Patrice Lefaucheur and
*Richard Nock**Robust Multiclass Ensemble Classifiers via Symmetric Functions***ICPR'06 - International Conference on Pattern Recognition**(*Hong-Kong, China*)

pp 136-139, 2006.*IAPR / IEEE press*.

- Svetlana Kiritchenko, Stan Matwin,
*Richard Nock*and Fazel Famili*Learning and Evaluation in the Presence of Class Hierarchies: Application to Text Categorization***AI'06 - Canadian Artificial Intelligence Conference**(*Québec, Canada*)

pp 395-406, 2006.*Springer Verlag*LNCS 4013.

- Frank Nielsen and
*Richard Nock**On the Smallest Enclosing Information Disk***CCCG'06 - Canadian Conference on Computational Geometry**(*Kingston, Canada*)

pp 131-134, 2006.

- Frank Nielsen and
*Richard Nock**ClickRemoval: interactive pinpoint image object removal***MM'05 - ACM Multimedia Conference**(*Singapore*)

pp 315-318, 2005.*ACM press*.

- Frank Nielsen and
*Richard Nock**Interactive Point-and-Click Segmentation for Object Removal in Digital Images***HCI'05 - IEEE International Conference on Human-Computer Interaction**(*Beijing, China*)

pp 131-140, 2005.*Springer Verlag*LNCS 3766.

- Pierre-Alain Laur,
*Richard Nock*, Jean-Emile Symphor and Pascal Poncelet*On the Estimation of Frequent Itemsets for Data Streams: Theory and Experiments***CIKM'05 - ACM International Conference on Information and Knowledge Management**(*Bremen, Allemagne*)

pp 327-328, 2005.*ACM Press*.

*Richard Nock*and Babak Esfandiari*Adaptive Filtering of Web Pages***PKDD'05 - European Conference on the Principle and Practice of Knowledge Discovery in Data Bases**(*Porto, Portugal*)

pp 634-642, 2005.*Springer Verlag*LNCS 3721.

*Richard Nock*and Frank Nielsen*Fitting the Smallest Enclosing Bregman Ball***ECML'05 - European Conference on Machine Learning**(*Porto, Portugal*)

pp 649-656, 2005.*Springer Verlag*LNCS 3720.

- Frank Nielsen and
*Richard Nock**Interactive Pinpoint Image Object Removal***CVPR'05 - IEEE International Conference on Computer Vision and Pattern Recognition**(*San Diego, USA*)

pp 1191 (plus video), 2005.*IEEE CS press*.

- Babak Esfandiari and
*Richard Nock**Adaptive Filtering of Advertisements on Web Pages***WWW'05 - International World Wide Web Conference**(*Chiba, Japan*)

pp 916-917, 2005.*ACM press*.

- Pierre-Alain Laur, Jean-Emile Symphor,
*Richard Nock*and Pascal Poncelet*Statistical Supports for Frequent Itemsets on Data Streams***MLDM'05 - IAPR International Conference on Machine Learning and Data Mining in Pattern Recognition**(*Leipzig, Allemagne*)

pp 395-404, 2005.*Springer Verlag*LNCS 3587.

- Jean-Christophe Janodet,
*Richard Nock*, Marc Sebban and Henri-Maxime Suchier*Boosting Grammatical Inference with Confidence Oracles***ICML'04 - International Conference on Machine Learning**(*Banff, Canada*)

pp 425-432, 2004.*Morgan Kaufmann / ACM Press*.

Related material (paper): here.

*Richard Nock*and Frank Nielsen*Grouping with Bias Revisited***CVPR'04 - IEEE International Conference on Computer Vision and Pattern Recognition**(*Washington DC, USA*)

pp 460-465, 2004.*IEEE CS press*.

Related material (paper): here.

Related material (page with sources, images, etc.): here.

- Frank Nielsen and
*Richard Nock**Approximating Smallest Enclosing Balls***ICCSA'04 - International Conference on Computational Science and its Applications**/

International Workshop on Computational Geometry and Applications (*Perugia, Italy*)

pp 147-157, 2004.*Springer Verlag*LNCS 3045.

*Richard Nock*and Frank Nielsen*An abstract Weighting Framework for Clustering Algorithms***SDM'04 - SIAM International Conference on Data Mining**(*Orlando, FL, USA*)

pp 200-209, 2004.*SIAM press*.

Related material (paper): here.

Related material (slides of the oral presentation): here.

*Richard Nock*and Frank Nielsen*Improving Clustering Algorithms through Constrained Convex Optimization***ICPR'04 - International Conference on Pattern Recognition**(*Cambridge, England*)

pp 557-560, 2004.*IAPR / IEEE press*.

*Richard Nock*and Vincent Pagé*Grouping with Bias as Distribution-free Mixture Model Estimation***ICPR'04 - International Conference on Pattern Recognition**(*Cambridge, England*)

pp 44-47, 2004.*IAPR / IEEE press*.

- Frank Nielsen and
*Richard Nock**Approximating Smallest Enclosing Disks***CCCG'04 - Canadian Conference on Computational Geometry**(*Montreal, Canada*)

pp 124-127, 2004.

Related material (paper): here.

- Frank Nielsen and
*Richard Nock**Small(est) enclosing Balls in Unbounded Dimensions***JCDCG'04 - Japanese Conference on Discrete and Computational Geometry**(*Tokyo, Japan*)*accepted*, 2004.

- Frank Nielsen and
*Richard Nock**On Region Merging: the Statistical Soundness of Fast Sorting, with Applications***CVPR'03 - IEEE International Conference on Computer Vision and Pattern Recognition**(*Madison, WI, USA*)

pp 19-26, 2003.*IEEE CS press*.

Related material (paper): here.

*Richard Nock*and Patrice Lefaucheur*A Robust Boosting Algorithm***ECML'02 - European Conference on Machine Learning**(*Helsinki, FI*)

pp 319-330, 2002.*Springer Verlag*LNCS 2430.

*Richard Nock**A Fast, Reliable Region-Merging like Approach Handling Occlusions***ICIP'02 - IEEE International Conference on Image Processing**(*Rochester, NY, USA*)*accepted*, 2002.*IEEE SP Press*.

*Richard Nock**Improving Noise Handling in Probabilistic Sorted Color Region Merging***ICASSP'02 - IEEE International Conference on Acoustics, Speech and Signal Processing**(*Orlando, FL, USA*)*accepted*, 2002.*IEEE SP Press*.

*Richard Nock**Fast and Reliable Color Region Merging inspired by Decision Tree Pruning***CVPR'01 - IEEE International Conference on Computer Vision and Pattern Recognition**(*Kauai, HI, USA*)

pp 271-276, 2001.*IEEE CS press*.

Related material (paper): here.

Related material (slides of the oral presentation): here.

Related material (page with sources, images, etc.): here.

- Marc Sebban,
*Richard Nock*and Stéphane Lallich*Boosting Neighborhood-Based Classifiers***ICML'01 - International Conference on Machine Learning**(*Williamstown, MA, USA*)

pp 505-512, 2001.*Morgan Kauffman*.

- Marc Sebban and
*Richard Nock**Improvement of Nearest-Neighbor Classifiers via Support Vector Machines***FLAIRS'01 - International FLAIRS Symposium**(*Key West, FL, USA*)

pp 113-117, 2001.*AAAI Press*.

- Marc Sebban and
*Richard Nock**Prototype Selection as an Information-preserving problem***ICML'00 - International Conference on Machine Learning**(*Stanford, CA, USA*)

pp 855-862, 2000.*Morgan Kauffman*.

- Marc Sebban and
*Richard Nock**Combining Feature and Prototype Pruning by Uncertainty Minimization***UAI'00 - International Conference on Uncertainty in Artificial Intelligence**(*Stanford, CA, USA*)

pp 533-540, 2000.*Morgan Kauffman*.

- Christophe Fiorio and
*Richard Nock**Sorted Region Merging to Maximize Test Reliability***ICIP'00 - IEEE International Conference on Image Processing**(*Vancouver, Canada*)

pp 808-811, 2000.*IEEE SP Press*.

*Richard Nock*and Marc Sebban*Sharper Bounds for the Hardness of Prototype and Feature Selection***ALT'00 - International Conference on Algorithmic Learning Theory**(*Sydney, Australia*)

pp 224-237, 2000.*Springer Verlag*LNCS 1968.

- Marc Sebban and
*Richard Nock**Contribution of Dataset Reduction Techniques to Tree-Simplification and Knowledge Discovery***PKDD'00 - European Conference on Principles and Practice of KDD**(*Lyon, France*)

pp 44-53, 2000.*Springer Verlag*LNAI 1910.

*Richard Nock*, Marc Sebban and Didier Bernard*A Symmetric Nearest-Neighbor Learning Rule***EWCBR2k - European Workshop on Case-Based Reasoning**(*Trento, Italy*)

pp 222-233, 2000.*Springer Verlag*LNCS 1898.

- Christophe Fiorio and
*Richard Nock**A Concentration-Based Adaptive Approach to Region Merging of Optimal Time and Space Complexities***BMVC'00 - British Machine Vision Conference**(*Bristol, England*)

pp 775-784, 2000.*SPR/IEE*.

- Marc Sebban and
*Richard Nock**Prototype Selection based on Information Theory***AI'00 - Canadian Artificial Intelligence Conference**(*Montreal, Canada*)

pp 90-101, 2000.*Springer Verlag*LNCS 1822.

*Richard Nock*and Marc Sebban*A New Prototype Weighting and Selection Scheme for Instance Based Learning Algorithms***FLAIRS'00 - International FLAIRS Symposium**(*Orlando, FL, USA*)

pp 71-75, 2000.*AAAI Press*.

*Richard Nock**Complexity in the Case against Accuracy: when Building one Function-Free Horn Clause is as Hard as Any***ALT'99 - International Conference on Algorithmic Learning Theory**(*Tokyo, Japan*)

pp 182-193, 1999.*Springer Verlag*LNAI 1720.

*Richard Nock*, Marc Sebban and Pascal Jappy*Experiments on a Representation-Independent "Top-down and Prune" Induction Scheme***PKDD'99 - European Conference on Principles and Practice of KDD**(*Prague, Czech Republic*)

pp 223-231 (long paper), 1999.*Springer Verlag*LNAI 1704.

- Marc Sebban and
*Richard Nock**Contribution of Boosting in Wrapper Models***PKDD'99 - European Conference on Principles and Practice of KDD**(*Prague, Czech Republic*)

pp 214-222 (long paper), 1999.*Springer Verlag*LNAI 1704.

*Richard Nock*and Pascal Jappy*A "Top-down and Prune" Induction Scheme for constrained Decision Committees***IDA'99 - International Symposium on Intelligent Data Analysis**(*Amsterdam, the Netherlands*)

pp 27-38 (long paper), 1999.*Springer Verlag*LNCS 1642.

- Christophe Fiorio and
*Richard Nock**Image segmentation using a Generic, Fast and Non-parametric approach***ICTAI'98 - IEEE International Conference on Tools with Artificial Intelligence**(*Taipei, Taiwan, ROC*)

pp 450-458, 1998.*IEEE CS Press*.

*Richard Nock*, Pascal Jappy and Jean Sallantin*Generalized Graph Colorability and Compressibility of Boolean Formulae***ISAAC'98 - International Symposium on Algorithms and Computation**(*Taejon, Korea*)

pp 237-246, 1998.*Springer Verlag*LNAI 1533.

Related material (paper): here.

*Richard Nock*and Pascal Jappy*Function-free Horn clauses are Hard to approximate***ILP'98 - International Conference on Inductive Logic Programming**(*Madison, WI, USA*)

pp 195-204, 1998.*Springer Verlag*LNAI 1446.

*Richard Nock*and Pascal Jappy*On the power of Decision Lists***ICML'98 International Conference on Machine Learning**(*Madison, WI, USA*)

pp 413-420, 1998.*Morgan Kaufmann*.

- Pascal Jappy and
*Richard Nock**PAC-learning Conceptual Graphs***ICCS'98 - International Conference on Conceptual Structures**(*Montpellier, France*)

pp 303-315, 1998.*Springer Verlag*LNCS 1453.

*Richard Nock*and Babak Esfandiari*Oracles and assistants : machine learning applied to network supervision***AI'98 - Canadian Artificial Intelligence Conference**(*Vancouver, Canada*)

pp 86-98, 1998.*Springer Verlag*LNCS 1418.

- Joel Quinqueton, Babak Esfandiari and
*Richard Nock**Chronicle learning and Agent-Oriented techniques for network management and supervision***IC2IN'97 - International Conference on Intelligent Networks and Intelligence in Networks**(*Paris, France*)*Invited paper*

pp 131-146, 1997.*Chapman & Hall*.

- Pascal Jappy,
*Richard Nock*and Olivier Gascuel*Negative Robust Learning Results for Horn clause programs***ICML'96 - International Conference on Machine Learning**(*Bari, Italie*)

pp 258-265, 1996.*Morgan Kaufmann*.

*Richard Nock*and Olivier Gascuel*On learning Decision Committees***ICML'95 - International Conference on Machine Learning**(*Tahoe City, CA, USA*)

pp 413-420, 1995.*Morgan Kaufmann*.

Related material (paper): here.

Since my recruitment, I have paid
particular attention to put the materials I use for my courses on
the web. As a matter of fact, I have taught quite a
*large* number of *various* courses. These courses are targeted at
various audiences (undergraduates and graduates).

These courses include-d-: Algebra, Analysis, Artificial Intelligence, Data Mining, Graph theory, Learning theory, Programming (C, Java), Statistics.

French materials used are available from the following link: http://www1.univ-ag.fr/~rnock/index-pedago.html. In part because they use non-public or copyrighted data, the access to these materials is restricted (contact me for the password-s-).

--- This page is certainly best viewed with
Firefox or Safari...