Inventor · disambiguated record
Pierre L. Dognin
Also filed as: DOGNIN PIERRE · DOGNIN PIERRE L
8 granted patents·3 pending applications·15 citations·filing 2010–2022
80Inventor score
Top patents by PatentIndex Score
11 records- 0177US11630989B2Mutual information neural estimation with Eta-trickIBM·Filed 2020·Granted Apr 18, 2023·2 cites·20 claims
- 0275US9626621B2Systems and methods for combining stochastic average gradient and hessian-free optimization for sequence training of deep neural networksIBM·Filed 2015·Granted Apr 18, 2017·3 cites·13 claims
- 0368US9483728B2Systems and methods for combining stochastic average gradient and hessian-free optimization for sequence training of deep neural networksIBM·Filed 2014·Granted Nov 1, 2016·4 cites·7 claims
- 0468US8972256B2System and method for dynamic noise adaptation for robust automatic speech recognitionRENNIE STEVEN J·Filed 2011·Granted Mar 3, 2015·5 cites·12 claims
- 0554US11829726B2Dual learning bridge between text and knowledge graphIBM·Filed 2021·Granted Nov 28, 2023·0 cites·20 claims
- 0652US2024070404A1Reinforced generation: reinforcement learning for text and knowledge graph bi-directional generation using pretrained language modelsIBM·Filed 2022·Application pending·0 cites
- 0751US2024013066A1Multi-stage knowledge graph construction using modelsIBM·Filed 2022·Application pending·0 cites
- 0849US8635067B2Model restructuring for client and server based automatic speech recognitionDOGNIN PIERRE·Filed 2010·Granted Jan 21, 2014·1 cites·24 claims
- 0947US12254390B2Wasserstein barycenter model ensemblingIBM·Filed 2019·Granted Mar 18, 2025·0 cites·20 claims
- 1045US9741341B2System and method for dynamic noise adaptation for robust automatic speech recognitionNUANCE COMMUNICATIONS INC·Filed 2015·Granted Aug 22, 2017·0 cites·20 claims
- 1136US2019147355A1Self-critical sequence training of multimodal systemsIBM·Filed 2017·Application pending·0 cites
Identity basis: PatentsView inventor disambiguation (2025Q4-odp release). How scoring works →