Zobrazeno 1 - 9
of 9
pro vyhledávání: '"Willette, Jeffrey"'
The transformer's context window is vital for tasks such as few-shot learning and conditional generation as it preserves previous tokens for active memory. However, as the context lengths increase, the computational costs grow quadratically, hinderin
Externí odkaz:
http://arxiv.org/abs/2406.17808
The Masked autoencoder (MAE) has drawn attention as a representative self-supervised approach for masked image modeling with vision transformers. However, even though MAE shows better generalization capability than fully supervised training from scra
Externí odkaz:
http://arxiv.org/abs/2405.18042
The transformer architecture has driven breakthroughs in recent years on tasks which require modeling pairwise relationships between sequential elements, as is the case in natural language understanding. However, long seqeuences pose a problem due to
Externí odkaz:
http://arxiv.org/abs/2310.01777
Masked image modeling (MIM) has become a popular strategy for self-supervised learning~(SSL) of visual representations with Vision Transformers. A representative MIM model, the masked auto-encoder (MAE), randomly masks a subset of image patches and r
Externí odkaz:
http://arxiv.org/abs/2210.02077
Recent work on mini-batch consistency (MBC) for set functions has brought attention to the need for sequentially processing and aggregating chunks of a partitioned set while guaranteeing the same output for all partitions. However, existing constrain
Externí odkaz:
http://arxiv.org/abs/2208.12401
Numerous recent works utilize bi-Lipschitz regularization of neural network layers to preserve relative distances between data instances in the feature spaces of each layer. This distance sensitivity with respect to the data aids in tasks such as unc
Externí odkaz:
http://arxiv.org/abs/2110.06381
Publikováno v:
35th Conference on Neural Information Processing Systems (NeurIPS 2021), Sydney, Australia
Most existing set encoding algorithms operate under the implicit assumption that all the set elements are accessible, and that there are ample computational and memory resources to load the set into memory during training and inference. However, both
Externí odkaz:
http://arxiv.org/abs/2103.01615
Neural networks have proven successful at learning from complex data distributions by acting as universal function approximators. However, they are often overconfident in their predictions, which leads to inaccurate and miscalibrated probabilistic pr
Externí odkaz:
http://arxiv.org/abs/2102.10803
Autor:
Wigfield, Donald C., Willette, Jeffrey E., MacKeen, Judy E., Perkins, Sherry L., Farant, Jean-Pierre
Publikováno v:
Journal of Analytical Toxicology; November 1982, Vol. 6 Issue: 6 p276-276, 1p
