Pith. sign in

REVIEW 19 cited by

A guide to convolution arithmetic for deep learning

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

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 1603.07285 v2 pith:BWCLKHQR submitted 2016-03-23 stat.ML cs.LGcs.NE

Vincent Dumoulin , Francesco Visin This is my paper · ORCID
classification stat.MLcs.LGcs.NE
keywords convolutionalguideshapedeeplayerslearningrelationshiptransposed
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

We introduce a guide to help deep learning practitioners understand and manipulate convolutional neural network architectures. The guide clarifies the relationship between various properties (input shape, kernel shape, zero padding, strides and output shape) of convolutional, pooling and transposed convolutional layers, as well as the relationship between convolutional and transposed convolutional layers. Relationships are derived for various cases, and are illustrated in order to make them intuitive.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 19 Pith papers

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

  1. Determining star formation histories and age-metallicity relations with convolutional neural networks

    astro-ph.GA 2026-05 unverdicted novelty 7.0 of 10

    A CNN with attention and shared latent space recovers SFHs and metallicities from spectro-photometric data with ~0.12 dex age and ~0.03 dex metallicity dispersion while running thousands of times faster than full spec...

  2. CodeJeNN: A simple C++ neural network generator for physics applications

    physics.comp-ph 2026-07 conditional novelty 6.0 of 10

    CodeJeNN auto-generates self-contained C++ code from trained Keras models, delivering 2–38× faster inference than Keras backends and enabling neural-network surrogates inside CFD solvers.

  3. Unlocking Realism and Interpretability in Wireless Channel Synthesis: A Physics-Guided Generative Approach

    eess.SP 2026-06 unverdicted novelty 6.0 of 10

    A physics-guided generative model synthesizes realistic, interpretable wireless channel matrices by linearizing a parametric geometric channel model and incorporating tensor decomposition for parameter flexibility.

  4. Physics-based Generative Models for Geometrically Consistent and Interpretable Wireless Channel Synthesis

    eess.SP 2025-05 conditional novelty 6.0 of 10

    A dictionary-based linearization of the geometric channel model lets a VAE generate physically valid, interpretable wireless channels with better distribution fidelity than prior generative baselines.

  5. A data-driven two-microphone method for in-situ sound absorption measurements

    cs.SD 2025-02 conditional novelty 6.0 of 10

    A 1D convolutional neural network maps two-microphone transfer functions from finite porous samples to the sound absorption coefficient of an equivalent infinite sample, validated with BEM simulations and free-field e...

  6. Biologically-inspired Semi-supervised Semantic Segmentation for Biomedical Imaging

    cs.CV 2024-12 conditional novelty 6.0 of 10

    A two-stage pipeline first learns features with Hebbian updates (no backprop), including new transpose-convolution rules, then fine-tunes on few labels, improving Dice on several medical segmentation benchmarks.

  7. A generalised novel loss function for computational fluid dynamics

    cs.LG 2024-11 conditional novelty 6.0 of 10

    A gradient-weighted mean squared error loss (GMSE) improves cGAN-generated CFD flow field quality and training speed on a submarine wake dataset compared to MSE.

  8. Deep learning networks for selection of persistent scatterer pixels in multi-temporal SAR interferometric processing

    eess.IV 2019-09 reject novelty 6.0 of 10

    New-application report: CNN and ConvLSTM networks trained on StaMPS labels select persistent scatterer pixels in Sentinel-1 interferograms, with reported validation accuracy of 93.50% for the LSTM variant.

  9. Jointly Aligning and Predicting Continuous Emotion Annotations

    cs.LG 2019-07 unverdicted novelty 6.0 of 10

    A multi-delay sinc network jointly aligns speech signals with delayed continuous emotion labels and predicts arousal/valence, claiming state-of-the-art speech-only results on RECOLA and SEWA.

  10. Aftab: A Comprehensive Benchmark of CNN Encoders and Advanced Value Functions in Parallelized Q-Networks

    cs.LG 2026-08 conditional novelty 5.0 of 10

    A composite CNN and multi-head value architecture, Aftab, improves buffer-free PQN on Atari-57 from 2.692 to 6.479 IQM human-normalized score, but its Procgen generalization advantage is fragile.

  11. Measuring and Predicting Where and When Pathologists Focus their Visual Attention while Grading Whole Slide Images of Cancer

    eess.IV 2025-08 unverdicted novelty 5.0 of 10

    The submission combines an attention prediction abstract with a federated learning body, so the claimed result cannot be evaluated from the provided material.

  12. Mapping Galaxy Images Across Ultraviolet, Visible and Infrared Bands Using Generative Deep Learning

    astro-ph.IM 2025-01 conditional novelty 5.0 of 10

    A supervised CNN with L1 plus SSIM loss translates galaxy images across ultraviolet, optical, and infrared bands, with high validation-set similarity on Illustris mocks and moderate similarity on DECaLS data.

  13. Compression of Higher Order Ambisonics with Multichannel RVQGAN

    cs.SD 2024-11 conditional novelty 5.0 of 10

    A multichannel RVQGAN with a covariance loss compresses 16-channel third-order Ambisonics to 16 kbps and outperforms Opus at 160 kbps in a MUSHRA listening test on ambient scenes.

  14. Learning Sensor Placement from Demonstration for UAV networks

    stat.AP 2019-09 conditional novelty 5.0 of 10

    The paper presents a CNN that learns a UAV placement utility function from expert demonstrations, achieving 75-91% agreement with expert locations and comparable quality of service in simulated missions.

  15. The model is the message: Lightweight convolutional autoencoders applied to noisy imaging data for planetary science and astrobiology

    astro-ph.EP 2025-07 conditional novelty 4.0 of 10

    A simple convolutional autoencoder reconstructs planetary images with up to 99% pixel loss, and the author argues its latent space could be a more efficient data product than raw imagery.

  16. Bidirectional Variational Autoencoders

    cs.LG 2025-05 conditional novelty 4.0 of 10

    A single shared-weight network that encodes on the forward pass and decodes on the backward pass can halve VAE parameters while keeping performance roughly on par with a standard VAE.

  17. Deep Learning for Time Series Forecasting: The Electric Load Case

    cs.LG 2019-07 unverdicted novelty 4.0 of 10

    Compares feedforward, recurrent, sequence-to-sequence and temporal convolutional neural networks for short-term electric load forecasting through experiments on two real datasets.

  18. Relative Layer-Wise Relevance Propagation: a more Robust Neural Networks eXplaination

    cs.LG 2025-01 reject novelty 2.0 of 10

    R-LRP is a hyperparameter-free LRP variant claimed to outperform prior methods, but it reduces to Input*Gradient and relies on an unproven conservation law.

  19. Understanding Deep Learning Techniques for Image Segmentation

    cs.CV 2019-07 unverdicted novelty 1.0 of 10

    A 2019 survey that categorizes and intuitively explains major deep learning techniques for image segmentation, progressing from classical methods to modern neural architectures.

Pith tools