Zobrazeno 1 - 10
of 186
pro vyhledávání: '"Huang, Bert"'
Supervised learning usually requires a large amount of labelled data. However, attaining ground-truth labels is costly for many tasks. Alternatively, weakly supervised methods learn with cheap weak signals that only approximately label some data. Man
Externí odkaz:
http://arxiv.org/abs/2302.09649
Publikováno v:
In Neural Networks February 2025 182
Autor:
Arachie, Chidubem, Huang, Bert
In many applications, training machine learning models involves using large amounts of human-annotated data. Obtaining precise labels for the data is expensive. Instead, training with weak supervision provides a low-cost alternative. We propose a nov
Externí odkaz:
http://arxiv.org/abs/2202.03987
Autor:
Huang, Bert I.1
Publikováno v:
Columbia Law Review. Apr2024, Vol. 124 Issue 3, p851-885. 35p.
Autor:
Arachie, Chidubem, Huang, Bert
Curation of large fully supervised datasets has become one of the major roadblocks for machine learning. Weak supervision provides an alternative to supervised learning by training with cheap, noisy, and possibly correlated labeling functions from va
Externí odkaz:
http://arxiv.org/abs/2009.07360
Autor:
Lu, You, Huang, Bert
Normalizing flows are deep generative models that allow efficient likelihood calculation and sampling. The core requirement for this advantage is that they are constructed using functions that can be efficiently inverted and for which the determinant
Externí odkaz:
http://arxiv.org/abs/2002.12229
Autor:
Huang, Bert I.
Publikováno v:
Columbia Law Review, 2022 May 01. 122(4), 90-109.
Externí odkaz:
https://www.jstor.org/stable/27136469
Autor:
Herbst, Alyssa, Huang, Bert
Annotating large unlabeled datasets can be a major bottleneck for machine learning applications. We introduce a scheme for inferring labels of unlabeled data at a fraction of the cost of labeling the entire dataset. Our scheme, bounded expectation of
Externí odkaz:
http://arxiv.org/abs/1906.07046
Autor:
Fan, Shuangfei, Huang, Bert
As a new approach to train generative models, \emph{generative adversarial networks} (GANs) have achieved considerable success in image generation. This framework has also recently been applied to data with graph structures. We propose labeled-graph
Externí odkaz:
http://arxiv.org/abs/1906.03220
Autor:
Arachie, Chidubem, Huang, Bert
The usage of machine learning models has grown substantially and is spreading into several application domains. A common need in using machine learning models is collecting the data required to train these models. In some cases, labeling a massive da
Externí odkaz:
http://arxiv.org/abs/1906.00512