HackerPapers

arXiv:1507.02672cs.NE

Semi-Supervised Learning with Ladder Networks

Antti Rasmus, Harri Valpola, Mikko Honkala, Mathias Berglund, Tapani Raiko

Abstract

We combine supervised learning with unsupervised learning in deep neural networks. The proposed model is trained to simultaneously minimize the sum of supervised and unsupervised cost functions by backpropagation, avoiding the need for layer-wise pre-training. Our work builds on the Ladder network proposed by Valpola (2015), which we extend by combining the model with supervision. We show that the resulting model reaches state-of-the-art performance in semi-supervised MNIST and CIFAR-10 classification, in addition to permutation-invariant MNIST classification with all labels.

arXiv comments
Revised denoising function, updated results, fixed typos
Journal reference
DOI
Subjects
cs.NENeural and Evolutionary Computing cs.LGMachine Learning stat.MLMachine Learning (Statistics)

On Hacker News

Linked by 4 identifiable commenters in 4 comments across 4 threads, from Sep 2015 to Mar 2025.

Comments linking this paper, per year.

Discussed in comments4 threads

  1. Ladder: Self-Improving LLMs Through Recursive Problem Decomposition 1 comment
  2. This AI Boom Will Also Bust 1 comment

    • This is somewhat tangental, but what do you think about Ladder Networks? (https://arxiv.org/abs/1507.02672, https://arxiv.org/abs/1511.06430) I was really impressed by their results, but I haven't seen it being applied successfully to other applications.

      ma2rtenView on HNParent context

  3. What is chemical intuition? 1 comment
  4. Towards universal neural nets: Gibbs machines and ACE 1 comment

    • Just to be abundantly clear - the ladder network [1] outscores this by a moderate margin, and is also actually SOTA without data augmentation. In my mind at least, adding data rotations in the latent space is still different than a fully connected model without data…

      kastnerkyleView on HNParent context

Submitted to Hacker News

Connected through shared HN threads, not inferred similarity or a recommendation.