Methods › General › Normalization › Weight Standardization
Weight Standardization
Introduced by Siyuan Qiao et al. in Micro-Batch Training with Batch-Channel Normalization and Weight Standardization
archive 2025-07-28 Description, source and code snippet are the archive's method entry.
Weight Standardization is a normalization technique that smooths the loss landscape by standardizing the weights in convolutional layers. Different from the previous normalization methods that focus on activations, WS considers the smoothing effects of weights more than just length-direction decoupling. Theoretically, WS reduces the Lipschitz constants of the loss and the gradients. Hence, WS smooths the loss landscape and improves training.
In Weight Standardization, instead of directly optimizing the loss ℒ on the original weights Ŵ, we reparameterize the weights Ŵ as a function of W, i.e. Ŵ=WS(W), and optimize the loss ℒ on W by SGD:
Ŵ = [ Ŵ_(i,j) | Ŵ_(i,j) = W_(i,j) - μ_(W_(i,·))σ_(W_(i,·)+ϵ)]
y = Ŵ*x
where
μ_(W_(i,·)) = 1I∑ⱼ₌₁ᴵW_(i, j), σ_(W_(i,·))=√(1I∑ᵢ₌₁ᴵ(W_(i,j) - μ_(W_(i,·)))²)
Similar to Batch Normalization, WS controls the first and second moments of the weights of each output channel individually in convolutional layers. Note that many initialization methods also initialize the weights in some similar ways. Different from those methods, WS standardizes the weights in a differentiable way which aims to normalize gradients during back-propagation. Note that we do not have any affine transformation on Ŵ. This is because we assume that normalization layers such as BN or GN will normalize this convolutional layer again.
In Syntology name not resolved (unknown_name). The lookup matched no method, paper or venue by this name; that asserts nothing about whether code exists.
Near matches the lookup lists, each a different name: Parametric Weights Standardization (PWS); Weight Mixup; weight scaling; Weight Weaving; Weight Recycling; weight operators; Weight Balancing; Filter-then-Weight.
Syntology's have check for this name (free, it answers coverage only), read 2026-10-05; ask it again live. A level is the served verification level of that one answer; nothing here is summed across methods.
Papers archive 2025-07-28
archive papers tagged: 12 · with a code link: 9 · where Syntology ran a sample: 5 (4 with a run with no instrument failure, 1 where every run was a failure of Syntology's instrument) Syntology
Show: all tagged papersonly where code ran (5 of 12 tagged: 4 with a run with no instrument failure, 1 where every run was a failure of Syntology's instrument)
Page 1 of 1: papers 1 to 12 of 12, newest first by the archive's date (ties by slug), in archive order.
Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code, as “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. It is per sample and not a correctness claim. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from.
-
Addressing Data Heterogeneity in Federated Learning with Adaptive Normalization-Free Feature Recalibration 2 Oct 2024 · 0 repositories · arXiv:2410.02006
-
Adaptive Gradient Regularization: A Faster and Generalizable Optimization Technique for Deep Neural Networks 24 Jul 2024 · 0 repositories · arXiv:2407.16944
-
Exploring Loss Functions for Time-based Training Strategy in Spiking Neural Networks 21 Sep 2023 · 1 repository
-
Rotational Equilibrium: How Weight Decay Balances Learning Across Neural Networks 26 May 2023 · 2 repositories · arXiv:2305.17212Syntology official (archive's flag): 5 ran · 5 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified (of 6 harvested samples)
-
Domain Adaptation and Active Learning for Fine-Grained Recognition in the Field of Biodiversity 22 Oct 2021 · 0 repositories · arXiv:2110.11778
-
Effect of Pre-Training Scale on Intra- and Inter-Domain Full and Few-Shot Transfer Learning for Natural and Medical X-Ray Chest Images 31 May 2021 · 1 repository · arXiv:2106.00116
-
ReCU: Reviving the Dead Weights in Binary Neural Networks 23 Mar 2021 · 3 repositories · arXiv:2103.12369Syntology community repositories only · 2 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified (of 3 harvested samples) · 3 pointer-only (licence)
-
Characterizing signal propagation to close the performance gap in unnormalized ResNets 21 Jan 2021 · 4 repositories · arXiv:2101.08692Syntology official (archive's flag): 1 ran · 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified (of 1 harvested sample)
-
Monocular Depth Estimation Using Laplacian Pyramid-Based Depth Residuals 8 Jan 2021 · 1 repository
-
Gradient Centralization: A New Optimization Technique for Deep Neural Networks 3 Apr 2020 · 8 repositories · arXiv:2004.01461Syntology community repositories only · 18 ran (of which 0 constructed an object rather than computing a result; 17 with no instrument failure: 1 honoured, 0 violated, 16 with no contract checked; 1 where Syntology's instrument failed) · 11 unverified (of 29 harvested samples) · 3 pointer-only (licence)
-
Big Transfer (BiT): General Visual Representation Learning 24 Dec 2019 · 9 repositories · arXiv:1912.11370Syntology community repositories only · 9 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 1 violated, 6 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified (of 10 harvested samples)
-
Micro-Batch Training with Batch-Channel Normalization and Weight Standardization 25 Mar 2019 · 7 repositories · arXiv:1903.10520
Tasks archive 2025-07-28
20 shown of 30 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.
Usage over time archive 2025-07-28
Components: the archive holds no method-to-method composition, so PwC's Components table cannot be rebuilt; the Papers list carries no Results column for the same reason (the archive does not join its leaderboard rows to method tags).
Categories archive 2025-07-28
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections