MNIST Dataset

The MNIST dataset (Lecun et al., 1998), which consists of 60,000 gray-scale images of handwritten digits. Each image has an accompanying label in {0, 1,..., 9}, and is stored as a 28 × 28 matrix that takes on values in [0, 255]. We first divide the entries of all the images by 255. Next, since there is no variation in the peripheral pixels of the images (Gallaugher & McNicholas, 2018), which violates model (1), we add an independent perturbation N (0, 0.01) to each element of the image. Finally, we vectorize each image to obtain a vector xi ∈ R784.

Data and Resources

Cite this as

Yann LeCun, L´eon Bottou, Yoshua Bengio, Patrick Haffner (2024). Dataset: MNIST Dataset. https://doi.org/10.57702/6w0eiklj

DOI retrieved: November 25, 2024

Additional Info

Field Value
Created November 25, 2024
Last update December 2, 2024
Defined In https://doi.org/10.48550/arXiv.1504.06787
Citation
  • https://doi.org/10.48550/arXiv.2302.02373
  • https://doi.org/10.48550/arXiv.1409.5185
  • https://doi.org/10.48550/arXiv.1912.13464
  • https://doi.org/10.1109/TNNLS.2019.2919948
  • https://doi.org/10.48550/arXiv.1711.04574
  • https://doi.org/10.48550/arXiv.1503.07906
  • https://doi.org/10.48550/arXiv.1909.08190
  • https://doi.org/10.48550/arXiv.1911.01575
  • https://doi.org/10.48550/arXiv.2306.06130
  • https://doi.org/10.48550/arXiv.2301.12033
  • https://doi.org/10.48550/arXiv.1910.12204
  • https://doi.org/10.48550/arXiv.2102.08184
  • https://doi.org/10.48550/arXiv.2405.12755
  • https://doi.org/10.48550/arXiv.2404.16208
  • https://doi.org/10.24963/ijcai.2020/590
  • https://doi.org/10.48550/arXiv.1910.02758
  • https://doi.org/10.48550/arXiv.1904.09601
  • https://doi.org/10.48550/arXiv.2212.13345
  • https://doi.org/10.48550/arXiv.2305.11650
  • https://doi.org/10.48550/arXiv.2209.11350
  • https://doi.org/10.48550/arXiv.2102.04426
  • https://doi.org/10.48550/arXiv.2205.15268
  • https://doi.org/10.48550/arXiv.2311.06372
  • https://doi.org/10.48550/arXiv.1805.06605
  • https://doi.org/10.48550/arXiv.2405.15727
  • https://doi.org/10.48550/arXiv.2304.02798
  • https://doi.org/10.48550/arXiv.2203.15267
  • https://doi.org/10.48550/arXiv.1811.11479
  • https://doi.org/10.48550/arXiv.1501.05494
  • https://doi.org/10.14722/ndss.2022.24058
  • https://doi.org/10.48550/arXiv.2207.04974
  • https://doi.org/10.48550/arXiv.2110.09164
  • https://doi.org/10.48550/arXiv.2007.01028
  • https://doi.org/10.48550/arXiv.2004.09388
  • https://doi.org/10.48550/arXiv.1611.02268
  • https://doi.org/10.1109/OJSP.2023.3349111
  • https://doi.org/10.48550/arXiv.2308.14930
  • https://doi.org/10.48550/arXiv.2303.16955
  • https://doi.org/10.48550/arXiv.2105.04801
  • https://doi.org/10.48550/arXiv.2107.01622
  • https://doi.org/10.48550/arXiv.1910.04958
  • https://doi.org/10.48550/arXiv.1709.09890
  • https://doi.org/10.48550/arXiv.1610.06249
  • https://doi.org/10.48550/arXiv.2010.00352
  • https://doi.org/10.48550/arXiv.2403.08124
  • https://doi.org/10.48550/arXiv.2401.15874
  • https://doi.org/10.48550/arXiv.2103.06797
  • https://doi.org/10.48550/arXiv.1807.05832
  • https://doi.org/10.48550/arXiv.2012.10547
  • https://doi.org/10.48550/arXiv.2102.02611
  • https://doi.org/10.1109/ICIP40778.2020.9190968
  • https://doi.org/10.48550/arXiv.2311.15603
  • https://doi.org/10.48550/arXiv.1602.02697
  • https://doi.org/10.48550/arXiv.2306.02174
  • https://doi.org/10.48550/arXiv.1903.07705
  • https://doi.org/10.48550/arXiv.2007.06661
  • https://doi.org/10.1016/j.knosys.2021.106816
  • https://doi.org/10.48550/arXiv.2103.16194
  • https://doi.org/10.48550/arXiv.2401.11694
Author Yann LeCun
More Authors
L´eon Bottou
Yoshua Bengio
Patrick Haffner
Homepage http://yann.lecun.com/exdb/mnist/