Pith. sign in

REVIEW 9 cited by

Deep Networks with Stochastic Depth

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1603.09382 v3 pith:ZE5L37EN submitted 2016-03-30 cs.LG cs.CVcs.NE

Deep Networks with Stochastic Depth

classification cs.LG cs.CVcs.NE
keywords networksdeeptrainingdepthlayerstesttimeerror
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Very deep convolutional networks with hundreds of layers have led to significant reductions in error on competitive benchmarks. Although the unmatched expressiveness of the many layers can be highly desirable at test time, training very deep networks comes with its own set of challenges. The gradients can vanish, the forward flow often diminishes, and the training time can be painfully slow. To address these problems, we propose stochastic depth, a training procedure that enables the seemingly contradictory setup to train short networks and use deep networks at test time. We start with very deep networks but during training, for each mini-batch, randomly drop a subset of layers and bypass them with the identity function. This simple approach complements the recent success of residual networks. It reduces training time substantially and improves the test error significantly on almost all data sets that we used for evaluation. With stochastic depth we can increase the depth of residual networks even beyond 1200 layers and still yield meaningful improvements in test error (4.91% on CIFAR-10).

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 9 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Subcritical Signal Propagation at Initialization in Normalization-Free Transformers

    cs.LG 2026-04 unverdicted novelty 7.0

    Transformers using elementwise tanh-like nonlinearities instead of LayerNorm show stretched-exponential APJN growth at large depth, indicating subcritical signal propagation unlike the power-law growth in pre-LayerNor...

  2. A Generalist Agent

    cs.AI 2022-05 accept novelty 7.0

    Gato is a multi-modal, multi-task, multi-embodiment generalist policy using one transformer network to handle text, vision, games, and robotics tasks.

  3. Wide Residual Networks

    cs.CV 2016-05 accept novelty 7.0

    Wide residual networks achieve higher accuracy and faster training than very deep thin residual networks by increasing width and decreasing depth, setting new state-of-the-art results on CIFAR, SVHN, and ImageNet.

  4. Stabilizing Extrapolation in Looped Transformers via Learned Stochastic Stopping

    cs.LG 2026-06 unverdicted novelty 6.0

    Stochastic loop counts during training of looped transformers reduce OOD variance on binary addition, Dyck-1, Unique Set and Copy tasks, with learned RL-Halting further improving the accuracy-stability trade-off.

  5. HealDA: Highlighting the importance of initial errors in end-to-end AI weather forecasts

    physics.ao-ph 2026-01 conditional novelty 6.0

    HealDA supplies ML-based initial conditions for AI weather models that produce forecasts trailing ERA5-initialized runs by less than one day of effective lead time, with the skill gap arising mainly from initial error size.

  6. Sharpness-Aware Minimization for Efficiently Improving Generalization

    cs.LG 2020-10 conditional novelty 6.0

    SAM solves a min-max problem to locate flat low-loss regions, improving generalization on CIFAR, ImageNet and label-noise tasks.

  7. mHC-SSM: Manifold-Constrained Hyper-Connections for State Space Language Models with Stream-Specialized Adapters

    cs.LG 2026-05 unverdicted novelty 5.0

    Manifold-constrained multi-stream mixing plus per-stream adapters improves SSM language model validation loss from 6.3507 to 6.1353 and perplexity from 572.91 to 461.88 on WikiText-2.

  8. A Wasserstein GAN-based climate scenario generator for risk management and insurance: the case of soil subsidence

    cs.LG 2026-04 unverdicted novelty 4.0

    A conditional Wasserstein GAN generates plausible future SWI drought trajectories for French insurance risk management under climate change.

  9. Deep learning applied to computational mechanics: A comprehensive review, state of the art, and the classics

    cs.LG 2022-12 unverdicted novelty 2.0

    A comprehensive review of deep learning techniques for computational mechanics, including LSTM for constitutive modeling, PINNs for PDE solving, optimizers, and kernel methods.