beta-VAE: Learning Basic Visual Concepts with a Constrained Variational Framework
Authors: Irina Higgins, Loic Matthey, Arka Pal, Christopher Burgess, Xavier Glorot, Matthew Botvinick, Shakir Mohamed, Alexander Lerchner
ICLR 2017 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We demonstrate that β-VAE with appropriately tuned β > 1 qualitatively outperforms VAE (β = 1), as well as state of the art unsupervised (Info GAN) and semi-supervised (DC-IGN) approaches to disentangled factor learning on a variety of datasets (celeb A, faces and chairs). Furthermore, we devise a protocol to quantitatively compare the degree of disentanglement learnt by different models, and show that our approach also significantly outperforms all baselines quantitatively. |
| Researcher Affiliation | Industry | Irina Higgins, Loic Matthey, Arka Pal, Christopher Burgess, Xavier Glorot, Matthew Botvinick, Shakir Mohamed, Alexander Lerchner Google Deep Mind {irinah,lmatthey,arkap,cpburgess,glorotx, botvinick,shakir,lerchner}@google.com |
| Pseudocode | No | The paper provides numbered steps for the disentanglement metric in Section 3 and detailed mathematical equations in Appendix A.4, but these are not formatted as explicit 'Pseudocode' or 'Algorithm' blocks. |
| Open Source Code | No | The paper does not provide any explicit statements about making the source code available or include any links to a code repository. |
| Open Datasets | Yes | We trained β-VAE (...) on a variety of datasets commonly used to evaluate disentangling performance of models: celeb A (Liu et al., 2015), chairs (Aubry et al., 2014) and faces (Paysan et al., 2009). |
| Dataset Splits | No | The paper mentions 'test samples' but does not specify training, validation, or test set splits, nor does it explicitly mention a validation set. |
| Hardware Specification | No | The paper does not specify any particular hardware used for running the experiments, such as specific GPU or CPU models. |
| Software Dependencies | No | The paper mentions optimizers like 'Adagrad' and 'Adam' and implicitly uses standard deep learning frameworks, but it does not provide specific version numbers for any software dependencies like programming languages, libraries (e.g., PyTorch, TensorFlow), or other tools used. |
| Experiment Setup | Yes | Table 1: Details of model architectures used in the paper. The models were trained using either adagrad (Duchi et al., 2011) or adam (Kingma & Ba, 2014) optimisers. ... Encoder Conv 32x4x4 (stride 2), 32x4x4 (stride 2), 64x4x4 (stride 2), 64x4x4 (stride 2), FC 256. Re LU activation. Latents 32 Decoder Deconv reverse of encoder. Re LU activation. Bernoulli. |