Výsledky vyhledávání - "Haffner, Patrick"

Report

The Role of Linguistic Priors in Measuring Compositional Generalization of Vision-Language Models

Autor: Wu, Chenwei, Li, Li Erran, Ermon, Stefano, Haffner, Patrick, Ge, Rong, Zhang, Zaiwei

Compositionality is a common property in many modalities including natural languages and images, but the compositional generalization of multi-modal models is not well-understood. In this paper, we identify two sources of visual-linguistic compositio

Externí odkaz: http://arxiv.org/abs/2310.02777

Zobrazit plný text záznamu

Report

Rapid Connectionist Speaker Adaptation

Autor: Witbrock, Michael, Haffner, Patrick

Publikováno v: ICASSP-92: 1992 IEEE International Conference on Acoustics, Speech, and Signal Processing, 1992, pp. 453-456 vol.1

We present SVCnet, a system for modelling speaker variability. Encoder Neural Networks specialized for each speech sound produce low dimensionality models of acoustical variation, and these models are further combined into an overall model of voice v

Externí odkaz: http://arxiv.org/abs/2211.08978

Zobrazit plný text záznamu

Report

Deep Clustering with Measure Propagation

Autor: Chen, Minhua, Jayakumar, Badrinath, Gopalakrishnan, Padmasundari, Huang, Qiming, Johnston, Michael, Haffner, Patrick

Deep models have improved state-of-the-art for both supervised and unsupervised learning. For example, deep embedded clustering (DEC) has greatly improved the unsupervised clustering performance, by using stacked autoencoders for representation learn

Externí odkaz: http://arxiv.org/abs/2104.08967

Zobrazit plný text záznamu

Report

Learning to Adapt by Minimizing Discrepancy

Autor: Ororbia II, Alexander G., Haffner, Patrick, Reitter, David, Giles, C. Lee

We explore whether useful temporal neural generative models can be learned from sequential data without back-propagation through time. We investigate the viability of a more neurocognitively-grounded approach in the context of unsupervised generative

Externí odkaz: http://arxiv.org/abs/1711.11542

Zobrazit plný text záznamu

Report

Accelerated Parallel Optimization Methods for Large Scale Machine Learning

Autor: Luo, Haipeng, Haffner, Patrick, Paiement, Jean-Francois

The growing amount of high dimensional data in different machine learning applications requires more efficient and scalable optimization algorithms. In this work, we consider combining two techniques, parallelism and Nesterov's acceleration, to desig

Externí odkaz: http://arxiv.org/abs/1411.6725

Zobrazit plný text záznamu

Apprentissage connexionniste et statistique global pour la reconnaissance de la parole

Autor: Haffner, Patrick.

Th. doct.--Signal et images--Paris--ENST, 1994.
Bibliogr. p. 181-192. Index. Résumé.

Externí odkaz: http://catalogue.bnf.fr/ark:/12148/cb35818523m

Zobrazit plný text záznamu

Report

Efficient Multiclass Implementations of L1-Regularized Maximum Entropy

Autor: Haffner, Patrick, Phillips, Steven, Schapire, Rob

This paper discusses the application of L1-regularized maximum entropy modeling or SL1-Max [9] to multiclass categorization problems. A new modification to the SL1-Max fast sequential learning algorithm is proposed to handle conditional distributions

Externí odkaz: http://arxiv.org/abs/cs/0506101

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání