Methods › Computer Vision › Likelihood-Based Generative Models › Beta-VAE
Beta-VAE
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))
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.
-
Causal Intervention Framework for Variational Auto Encoder Mechanistic Interpretability 6 May 2025 · 0 repositories · arXiv:2505.03530
-
Alternatives of Unsupervised Representations of Variables on the Latent Space 26 Oct 2024 · 0 repositories · arXiv:2410.20172
-
Comparison of Autoencoder Encodings for ECG Representation in Downstream Prediction Tasks 3 Oct 2024 · 0 repositories · arXiv:2410.02937
-
Lewis's Signaling Game as beta-VAE For Natural Word Lengths and Segments 8 Nov 2023 · 1 repository · arXiv:2311.04453
-
From Conceptual Spaces to Quantum Concepts: Formalising and Learning Structured Conceptual Models 6 Nov 2023 · 0 repositories · arXiv:2401.08585
-
Learning minimal representations of stochastic processes with variational autoencoders 21 Jul 2023 · 1 repository · arXiv:2307.11608
-
Impact of Disentanglement on Pruning Neural Networks 19 Jul 2023 · 0 repositories · arXiv:2307.09994
-
Identification of Rare Cortical Folding Patterns using Unsupervised Deep Learning 29 Nov 2022 · 1 repository · arXiv:2211.16213
-
Analysis of Master Vein Attacks on Finger Vein Recognition Systems 18 Oct 2022 · 0 repositories · arXiv:2210.10667
-
TopicVAE: Topic-aware Disentanglement Representation Learning for Enhanced Recommendation 10 Oct 2022 · 1 repository
-
The Conceptual VAE 21 Mar 2022 · 0 repositories · arXiv:2203.11216
-
Designing Complex Experiments by Applying Unsupervised Machine Learning 29 Sep 2021 · 0 repositories · arXiv:2110.01458
-
Be More Active! Understanding the Differences between Mean and Sampled Representations of Variational Autoencoders 26 Sep 2021 · 1 repository · arXiv:2109.12679
-
Finding simplicity: unsupervised discovery of features, patterns, and order parameters via shift-invariant variational autoencoders 23 Jun 2021 · 1 repository · arXiv:2106.12472
-
Recovering Barabási-Albert Parameters of Graphs through Disentanglement 3 May 2021 · 1 repository · arXiv:2105.00997
-
Learning Stable Representations with Full Encoder 25 Mar 2021 · 0 repositories · arXiv:2103.14082
-
Dynamic β-VAEs for quantifying biodiversity by clustering optically recorded insect signals 10 Feb 2021 · 1 repository · arXiv:2102.05526
-
Autoencoding Slow Representations for Semi-supervised Data Efficient Regression 11 Dec 2020 · 0 repositories · arXiv:2012.06279
-
AI Discovering a Coordinate System of Chemical Elements: Dual Representation by Variational Autoencoders 24 Nov 2020 · 0 repositories · arXiv:2011.12090
-
Unsupervised anomaly localization using VAE and beta-VAE 19 May 2020 · 0 repositories · arXiv:2005.10686
-
AI Giving Back to Statistics? Discovery of the Coordinate System of Univariate Distributions by Beta Variational Autoencoder 6 Apr 2020 · 0 repositories · arXiv:2004.02687
-
Variational Learning with Disentanglement-PyTorch 11 Dec 2019 · 1 repository · arXiv:1912.05184Syntology ran 1 of 1 samples · 0 unverified · 1 pointer-only (licence)
-
Information bottleneck through variational glasses 2 Dec 2019 · 0 repositories · arXiv:1912.00830
-
Flatsomatic: A Method for Compression of Somatic Mutation Profiles in Cancer 27 Nov 2019 · 0 repositories · arXiv:1911.13259
-
A Preliminary Study of Disentanglement With Insights on the Inadequacy of Metrics 26 Nov 2019 · 0 repositories · arXiv:1911.11791
-
Disentangling Interpretable Generative Parameters of Random and Real-World Graphs 12 Oct 2019 · 0 repositories · arXiv:1910.05639
-
IB-GAN: Disentangled Representation Learning with Information Bottleneck GAN 1 May 2019 · 2 repositories
-
ISA-VAE: Independent Subspace Analysis with Variational Autoencoders 1 May 2019 · 0 repositories
-
The Information-Autoencoding Family: A Lagrangian Perspective on Latent Variable Generative Modeling 1 Jan 2018 · 0 repositories
-
beta-VAE: Learning Basic Visual Concepts with a Constrained Variational Framework 26 Apr 2017 · 6 repositories
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.
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