We lifted 8 functions out of this paper's own repositories and ran 4 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| oskopek/mvae | canonical | 4 of 8 |
| Function | Status | Where it lives |
|---|---|---|
| all_models | Ran | oskopek/mvae/mt/mvae/all_models.py code served (permissive licence) · get_code("e1fe823ad0ece20a") |
| canonical_name | Ran | oskopek/mvae/mt/mvae/all_models.py code served (permissive licence) · get_code("7a3010c14a89f246") |
| diff | Ran | oskopek/mvae/mt/mvae/distributions/ars.py code served (permissive licence) · get_code("81544b61f125ae3d") |
| log_sum_exp_signs | Ran | oskopek/mvae/mt/mvae/distributions/pvae_utils.py code served (permissive licence) · get_code("e89a04d86cc8eb8f") |
| cdf_r | Not yet run | oskopek/mvae/mt/mvae/distributions/hyperbolic_radius.py code served (permissive licence) · get_code("360bfcf03cfbec8a") |
| grad_cdf_value_scale | Not yet run | oskopek/mvae/mt/mvae/distributions/hyperbolic_radius.py code served (permissive licence) · get_code("c7ec5f7b40878d0f") |
| parse_component_str | Not yet run | oskopek/mvae/mt/mvae/utils.py code served (permissive licence) · get_code("5a61d0a71b89572c") |
| rexpand | Not yet run | oskopek/mvae/mt/mvae/distributions/pvae_utils.py code served (permissive licence) · get_code("f09eaf2b8361cf4b") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Euclidean geometry has historically been the typical "workhorse" for machine learning applications due to its power and simplicity. However, it has recently been shown that geometric spaces with constant non-zero curvature improve representations and performance on a variety of data types and downstream tasks. Consequently, generative models like Variational Autoencoders (VAEs) have been successfully generalized to elliptical and hyperbolic latent spaces. While these approaches work well on data with particular kinds of biases e.g. tree-like data for a hyperbolic VAE, there exists no generic approach unifying and leveraging all three models. We develop a Mixed-curvature Variational Autoencoder, an efficient way to train a VAE whose latent space is a product of constant curvature Riemannian manifolds, where the per-component curvature is fixed or learnable. This generalizes the Euclidean VAE to curved latent spaces and recovers it when curvatures of all latent space components go to 0.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("1911.08411")
get_code_for_paper("1911.08411")
have("1911.08411")
Connect an agent — have() is free.