Methods › Computer Vision › Likelihood-Based Generative Models › Beta-VAE

Beta-VAE

30 papers tagged archive 2025-07-28

Introduced by Irina Higgins et al. in beta-VAE: Learning Basic Visual Concepts with a Constrained Variational Framework

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

Beta-VAE is a type of variational autoencoder that seeks to discover disentangled latent factors. It modifies VAEs with an adjustable hyperparameter β that balances latent channel capacity and independence constraints with reconstruction accuracy. The idea is to maximize the probability of generating the real data while keeping the distance between the real and estimated distributions small, under a threshold ϵ. We can use the Kuhn-Tucker conditions to write this as a single equation:

ℱ(θ, ϕ, β; 𝐱, 𝐳) = 𝔼_(q_ϕ(𝐳|𝐱))[logp_θ(𝐱|𝐳)] - β[D_(KL)(logq_θ(𝐳|𝐱)||p(𝐳)) - ϵ]

where the KKT multiplier β is the regularization coefficient that constrains the capacity of the latent channel 𝐳 and puts implicit independence pressure on the learnt posterior due to the isotropic nature of the Gaussian prior p(𝐳).

We write this again using the complementary slackness assumption to get the Beta-VAE formulation:

ℱ(θ, ϕ, β; 𝐱, 𝐳) ≥ℒ(θ, ϕ, β; 𝐱, 𝐳) = 𝔼_(q_ϕ(𝐳|𝐱))[logp_θ(𝐱|𝐳)] - β{D_(KL)(logq_θ(𝐳|𝐱)||p(𝐳))

PaperSourceSee Code · AntixK/PyTorch-VAE

Papers archive 2025-07-28

30 shown of 30, newest first. Repository counts are the archive's code-links table. A Syntology line states what Syntology ran from that paper's harvested code; it is per sample and not a correctness claim.

Tasks archive 2025-07-28

20 shown of 22 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
Disentanglement14
Representation Learning5
Clustering2
Data Compression2
Graph Generation2
Variational Inference2
Anomaly Detection1
Anomaly Localization1
BIG-bench Machine Learning1
Data Visualization1
Decoder1
Deep Learning1
Finger Vein Recognition1
Graph Embedding1
Graph Neural Network1
Graph Representation Learning1
Model Compression1
Novelty Detection1
Recommendation Systems1
Scheduling1

Usage over time archive 2025-07-28

Papers per year tagged with Beta-VAE: 2017 to 2025, peak 7 7 0 2017: 1 paper 2017 2018: 1 paper 2018 2019: 7 papers 2019 2020: 4 papers 2020 2021: 6 papers 2021 2022: 4 papers 2022 2023: 4 papers 2023 2024: 2 papers 2024 2025: 1 paper 2025
Papers per year the archive tags with this method, by the paper's archive date (30 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

Likelihood-Based Generative ModelsGenerative Models

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