Csi: Novelty detection via contrastive learning on distributionally shifted instances

J Tack, S Mo, J Jeong, J Shin - Advances in neural …, 2020 - proceedings.neurips.cc
Novelty detection, ie, identifying whether a given sample is drawn from outside the training
distribution, is essential for reliable machine learning. To this end, there have been many …
被引用次数：608 相关文章

CSI: novelty detection via contrastive learning on distributionally shifted instances

J Tack, S Mo, J Jeong, J Shin - … of the 34th International Conference on …, 2020 - dl.acm.org
Novelty detection, ie, identifying whether a given sample is drawn from outside the training
distribution, is essential for reliable machine learning. To this end, there have been many …

CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instances

J Tack, S Mo, J Jeong, J Shin - 34th Conference on Neural …, 2020 - koasas.kaist.ac.kr
Novelty detection, ie, identifying whether a given sample is drawn from outside the training
distribution, is essential for reliable machine learning. To this end, there have been many …

CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instances!

J Tack, S Mo, J Jeong, J Shin - jihoontack.github.io
TL;DR. We propose a novel contrastive learning scheme for out-of-distribution (OOD) detection,
which contrasts hard (distributio Page 1 CSI: Novelty Detection via Contrastive Learning on …
[PDF] nips.cc

[PDF][PDF] CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instances

J Tack, S Mo, J Jeong, J Shin - proceedings.nips.cc
Novelty detection, ie, identifying whether a given sample is drawn from outside the training
distribution, is essential for reliable machine learning. To this end, there have been many …

CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instances

J Tack, S Mo, J Jeong, J Shin - Advances in Neural …, 2020 - proceedings.neurips.cc
Novelty detection, ie, identifying whether a given sample is drawn from outside the training
distribution, is essential for reliable machine learning. To this end, there have been many …

CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instances!

J Tack, S Mo, J Jeong, J Shin - jh-jeong.github.io
TL;DR. We propose a novel contrastive learning scheme for out-of-distribution (OOD) detection,
which contrasts hard (distributio Page 1 CSI: Novelty Detection via Contrastive Learning on …

CSI: Novelty detection via contrastive learning on distributionally shifted instances

J Tack, S Mo, J Jeong, J Shin - Advances in Neural Information …, 2020 - pure.kaist.ac.kr
Novelty detection, ie, identifying whether a given sample is drawn from outside the training
distribution, is essential for reliable machine learning. To this end, there have been many …

CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instances

J Tack, S Mo, J Jeong, J Shin - arXiv e-prints, 2020 - ui.adsabs.harvard.edu
Novelty detection, ie, identifying whether a given sample is drawn from outside the training
distribution, is essential for reliable machine learning. To this end, there have been many …

CSI: Novelty Detection via Contrastive Learning on Distributionally Shifted Instances

J Tack, S Mo, J Jeong, J Shin - arXiv preprint arXiv:2007.08176, 2020 - arxiv.org
Novelty detection, ie, identifying whether a given sample is drawn from outside the training
distribution, is essential for reliable machine learning. To this end, there have been many …