Methods › General › Normalization › Conditional Batch Normalization

Conditional Batch Normalization

145 papers tagged archive 2025-07-28

Introduced by Harm de Vries et al. in Modulating early visual processing by language

archive 2025-07-28 Description, source and code snippet are the archive's method entry.

Conditional Batch Normalization (CBN) is a class-conditional variant of batch normalization. The key idea is to predict the γ and β of the batch normalization from an embedding - e.g. a language embedding in VQA. CBN enables the linguistic embedding to manipulate entire feature maps by scaling them up or down, negating them, or shutting them off. CBN has also been used in GANs to allow class information to affect the batch normalization parameters.

Consider a single convolutional layer with batch normalization module BN(F_(i,c,h,w)|γ_c, β_c) for which pretrained scalars γ_c and β_c are available. We would like to directly predict these affine scaling parameters from, e.g., a language embedding e_q. When starting the training procedure, these parameters must be close to the pretrained values to recover the original ResNet model as a poor initialization could significantly deteriorate performance. Unfortunately, it is difficult to initialize a network to output the pretrained γ and β. For these reasons, the authors propose to predict a change δβ_c and δγ_c on the frozen original scalars, for which it is straightforward to initialize a neural network to produce an output with zero-mean and small variance.

The authors use a one-hidden-layer MLP to predict these deltas from a question embedding e_q for all feature maps within the layer:

Δβ= MLP(e_q)

Δγ= MLP(e_q)

So, given a feature map with C channels, these MLPs output a vector of size C. We then add these predictions to the β and γ parameters:

β̂_c = β_c + Δβ_c

γ̂_c = γ_c + Δγ_c

Finally, these updated β̂ and γ̂ are used as parameters for the batch normalization: BN(F_(i,c,h,w)|γ̂_̂ĉ, β̂_̂ĉ). The authors freeze all ResNet parameters, including γ and β, during training. A ResNet consists of four stages of computation, each subdivided in several residual blocks. In each block, the authors apply CBN to the three convolutional layers.

PaperSourceSee Code · ap229997/Conditional-Batch-Norm

In Syntology for a different name, 'Batch Normalization', about another paper, arXiv:1502.03167: repository known, no verified implementation (declared_code; repositories: 5 listed of 66 in the graph, strongest link: mentioned on GitHub). The lookup matched a different name in Syntology's graph, 'Batch Normalization' (resolution: token_match); this answer is about that name. About another paper: this answer is about arXiv:1502.03167, while the archive credits arXiv:1707.00683.

Code Syntology holds under other names, as the lookup lists them (code_nearby); not a complete list, and none of them is this method.

Origin paper of this answer, arXiv:1502.03167 (its page on this site · its page on Syntology): Syntology's run record for its harvested code, 19 ran (of which 5 constructed an object rather than computing a result; 14 with no instrument failure: 0 honoured, 0 violated, 14 with no contract checked; 5 where Syntology's instrument failed) · 2 unverified (of 21 harvested samples) · 5 pointer-only (licence). The instrument figure counts failures of Syntology's instrument, not of the code. Ran means executed on a synthesized input, not that the code is correct.

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.

What each level means, quoted from Syntology's server card: V0 “imports”; V1 “executes against a synthesized fixture and returns a value of the recorded type”; V2 “two implementations GENERATED from the paper's signature and spec agree numerically (1e-6) on fixture inputs” “neither of the two ever read the authors' own code, so V2 measures how unambiguously the paper specifies the method, not fidelity to the authors' implementation (measured 2026-09-11, commit 75eaa77: cross_checked_by is a second LLM on 2,628 of 2,831 served samples)”; V3 “paper-derived property tests pass on both of those same two generations.” The card adds: “No rung on this ladder compares the served code against the authors' implementation, so treat every level as evidence about runnability and internal agreement, not about matching the paper's results.” A level quoted in two parts is one definition the card writes as two clauses joined by a double hyphen, each clause quoted on its own (how these levels are quoted).

Papers archive 2025-07-28

archive papers tagged: 145 · with a code link: 79 · where Syntology ran a sample: 26 (22 with a run with no instrument failure, 4 where every run was a failure of Syntology's instrument) Syntology

Show: all tagged papersonly where code ran (26 of 145 tagged: 22 with a run with no instrument failure, 4 where every run was a failure of Syntology's instrument)

Page 1 of 2: papers 1 to 100 of 145, 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.

Tasks archive 2025-07-28

20 shown of 145 tasks the archive attaches to papers tagged with this method, by distinct papers. A task without a page in the catalog is plain text.

TaskPapers
Image Generation42
Conditional Image Generation16
Generative Adversarial Network12
reinforcement-learning9
Multi-agent Reinforcement Learning7
Reinforcement Learning7
Data Augmentation6
Reinforcement Learning (RL)6
Super-Resolution6
Vocal Bursts Intensity Prediction6
Decision Making5
Transfer Learning5
Unconditional Image Generation5
Attribute4
Clustering4
Denoising4
Diversity4
Object4
Benchmarking3
Decoder3

Usage over time archive 2025-07-28

Papers per year tagged with Conditional Batch Normalization: 2017 to 2024, peak 38 38 0 2017: 2 papers 2017 2018: 4 papers 2018 2019: 11 papers 2019 2020: 31 papers 2020 2021: 21 papers 2021 2022: 35 papers 2022 2023: 38 papers 2023 2024: 3 papers 2024
Papers per year the archive tags with this method, by the paper's archive date (145 dated). Bars are counts, not a trend claim.

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

Normalization

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