 
    Publications
                    
                    
                    7
                
                            
                    
                    
                    4
                
                            
                    
                    
                    4
                
                            
                    
                    
                    4
                
                            
                    
                    
                    3
                
                            
                    
                    
                    3
                
                            
                    
                    
                    3
                
                            
                    
                    
                    3
                
                            
                    
                    
                    2
                
                            
                    
                    
                    2
                
                            
                    
                    
                    2
                
                            
                    
                    
                    2
                
                            
                    
                    
                    2
                
                            
                    
                    
                    2
                
                            
                    
                    
                    2
                
                            
                    
                    
                    2
                
                            
                    
                    
                    2
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                            
                    
                    
                    1
                
                    Publications
|  | PETRA: Parallel End-to-end Training with Reversible Architectures2025Pré-publication, Document de travail
                    
        
        
            
                
                    hal-04594647
                                            v2 | 
|  | Decentralized Asynchronous Optimization with DADAO allows Decoupling and Acceleration2025Pré-publication, Document de travail
                    
        
        
            
                
                    hal-05270499
                                            v1 | 
|  | Vectorizing string entries for data processing on tables: when are larger language models better?2023Pré-publication, Document de travail
                    
        
        
            
                
                    hal-04345931
                                            v1 | 
|  | Decoupled Greedy Learning of CNNs for Synchronous and Asynchronous Distributed Learning2021Pré-publication, Document de travail
                    
        
        
            
                
                    hal-03247753
                                            v1 | 
|  | Interferometric Graph Transform for Community Labeling2021Pré-publication, Document de travail
                    
        
        
            
                
                    hal-03247781
                                            v1 | 
|  | Nonlinear Acceleration of Deep Neural Networks2018Pré-publication, Document de travail
                    
        
        
            
                
                    hal-01799269
                                            v1 | 
|  | ACCO: Accumulate While You Communicate for Communication-Overlapped Sharded LLM TrainingThe Thirty-ninth Annual Conference on Neural Information Processing Systems, Dec 2025, San diego (Californie), United StatesCommunication dans un congrès
                    
        
        
            
                
                    hal-04592562
                                            v3 | 
|  | $\textbf{A}^2\textbf{CiD}^2$: Accelerating Asynchronous Communication in Decentralized Deep LearningThirty-seventh Conference on Neural Information Processing Systems, Dec 2023, New Orleans, United StatesCommunication dans un congrès
                    
        
        
            
                
                    hal-04124318
                                            v2 | 
|  | Can Forward Gradient Match Backpropagation?Fortieth International Conference on Machine Learning, Jul 2023, Honolulu (Hawaii), USA, United StatesCommunication dans un congrès
                    
        
        
            
                
                    hal-04119829
                                            v1 | 
|  | Preventing Dimensional Collapse in Contrastive Local Learning with SubsamplingICML 2023 Workshop on Localized Learning (LLW), Jul 2023, Honolulu (Hawaii), USA, United StatesCommunication dans un congrès
                    
        
        
            
                
                    hal-04156218
                                            v1 | 
|  | DADAO: Decoupled Accelerated Decentralized Asynchronous OptimizationInternational Conference on Machine Learning, Jul 2023, Honolulu, United StatesCommunication dans un congrès
                    
        
        
            
                
                    hal-03737694
                                            v3 | 
|  | Why do tree-based models still outperform deep learning on typical tabular data?36th Conference on Neural Information Processing Systems (NeurIPS 2022) Track on Datasets and Benchmarks, Nov 2022, New Orleans, United StatesCommunication dans un congrès
                    
        
        
            
                
                    hal-03723551
                                            v3 | 
|  | On Non-Linear operators for Geometric Deep LearningConference on Neural Information Processing Systems (Neurips), Dec 2022, New Orleans, United States. ⟨10.48550/arXiv.2207.03485⟩Communication dans un congrès
                    
        
        
            
                
                    hal-03711864
                                            v2 | 
|  | Deep Reinforcement Learning for L3 Slice Localization in Sarcopenia AssessmentMachine Learning in Medical Imaging @MICCAI, Sep 2021, Strasbourg, France. pp.317-326, ⟨10.1007/978-3-030-87589-3_33⟩Communication dans un congrès
                    
        
        
            
                
                    hal-03532954
                                            v1 | 
|  | The Unreasonable Effectiveness of Patches in Deep Convolutional Kernels MethodsInternational Conference on Learning Representation (ICLR 2021), 2021, Vienna (online), AustriaCommunication dans un congrès
                    
        
        
            
                
                    hal-03114389
                                            v1 | 
|  | Low-Rank Projections of GCNs LaplacianICLR 2021 Workshop GTRL, May 2021, Online, FranceCommunication dans un congrès
                    
        
        
            
                
                    hal-03248056
                                            v1 | 
|  | Interferometric Graph Transform: a Deep Unsupervised Graph Representation37th International Conference on Machine Learning (ICML 2020), Jul 2020, Online, AustriaCommunication dans un congrès
                    
        
        
            
                
                    hal-02862516
                                            v1 | 
|  | Decoupled Greedy Learning of CNNsInternational Conference on Machine Learning, Jul 2020, Vienna (virtual), Austria. pp.5368-5377Communication dans un congrès
                    
        
        
            
                
                    hal-02945327
                                            v1 | 
|  | On Lazy Training in Differentiable ProgrammingNeurIPS 2019 - 33rd Conference on Neural Information Processing Systems, Dec 2019, Vancouver, Canada. pp.2937-2947Communication dans un congrès
                    
        
        
            
                
                    hal-01945578
                                            v6 | 
|  | Greedy Layerwise Learning Can Scale to ImageNetICML 2019 - 36th International Conference on Machine Learning, Jun 2019, Long Beach, CA, United StatesCommunication dans un congrès
                    
        
        
            
                
                    hal-02119398
                                            v1 | 
|  | Compressing the Input for CNNs with the First-Order Scattering TransformECCV 2018 - European Conference on Computer Vision, Sep 2018, Munich, GermanyCommunication dans un congrès
                    
        
        
            
                
                    hal-01850921
                                            v1 | 
|  | i-RevNet: Deep Invertible NetworksICLR 2018 - International Conference on Learning Representations, Apr 2018, Vancouver, CanadaCommunication dans un congrès
                    
        
        
            
                
                    hal-01712808
                                            v1 | 
|  | Nonlinear Acceleration of CNNsICLR Workshop track, Apr 2018, Vancouver, CanadaCommunication dans un congrès
                    
        
        
            
                
                    hal-01805251
                                            v3 | 
|  | Scaling the Scattering Transform: Deep Hybrid NetworksInternational Conference on Computer Vision (ICCV), Oct 2017, Venice, ItalyCommunication dans un congrès
                    
        
        
            
                
                    hal-01495734
                                            v2 | 
|  | Contributions to Local, Asynchronous and Decentralized Learning, and to Geometric Deep LearningArtificial Intelligence [cs.AI]. Sorbonne Université, 2023HDR
                    
        
        
            
                
                    tel-04334118
                                            v1 | 
|  | Guiding The Last Layer in Federated Learning with Pre-Trained ModelsNeurips, In pressArticle dans une revue
                    
        
        
            
                
                    hal-04262471
                                            v1 | 
|  | Gradient Masked Averaging for Federated LearningTransactions on Machine Learning Research Journal, 2022Article dans une revue
                    
        
        
            
                
                    hal-04262447
                                            v1 | 
|  | Kymatio: Scattering Transforms in PythonJournal of Machine Learning Research, 2020, 21 (60), pp.1-6Article dans une revue
                    
        
        
            
                
                    hal-02945354
                                            v1 | 
|  | Scattering Networks for Hybrid Representation LearningIEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, pp.11. ⟨10.1109/TPAMI.2018.2855738⟩Article dans une revue
                    
        
        
            
                
                    hal-01837587
                                            v1 | 
|  | An Analysis of the SURF MethodImage Processing On Line, 2015, 5, pp.176 - 218. ⟨10.5201/ipol.2015.69⟩Article dans une revue
                    
        
        
            
                
                    hal-01657161
                                            v1 | 
|  | Analyzing and introducing structures in deep convolutional neural networksComputer Vision and Pattern Recognition [cs.CV]. Université Paris sciences et lettres, 2017. English. ⟨NNT : 2017PSLEE060⟩Thèse
                    
        
        
            
                
                    tel-03917151
                                            v2 | 
                Chargement...
            
            Chargement...