SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2405.16763 · 2024

Transport of Algebraic Structure to Latent Embeddings

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 10 functions out of this paper's own repositories and ran 9 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.

RepositoryRoleRan
spfrommer/latent_algebras canonical 9 of 10
FunctionStatusWhere it lives
add_click_options Ran spfrommer/latent_algebras/latalg/utils/cmd_utils.py
pointer only (licence: NONE) · get_code("6c055a4b53038c84")
change_extension Ran spfrommer/latent_algebras/latalg/utils/file_utils.py
pointer only (licence: NONE) · get_code("cc1302fa5d7c53c3")
create_mlp_from_params Ran spfrommer/latent_algebras/latalg/utils/model_utils.py
pointer only (licence: NONE) · get_code("fe839e5b459df206")
files_with_extension Ran spfrommer/latent_algebras/latalg/utils/file_utils.py
pointer only (licence: NONE) · get_code("3fcdad9780a321ef")
flatten_mlp_params Ran spfrommer/latent_algebras/lib/inr2vec/utils.py
pointer only (licence: NONE) · get_code("c6fcd3c4829c610c")
get_mlps_batched_params Ran spfrommer/latent_algebras/lib/inr2vec/utils.py
pointer only (licence: NONE) · get_code("2e91ed7aa3fa38b2")
remove_extension Ran spfrommer/latent_algebras/latalg/utils/file_utils.py
pointer only (licence: NONE) · get_code("9e0c05cabd0e3ce5")
symbolize Ran spfrommer/latent_algebras/latalg/main/algebra.py
pointer only (licence: NONE) · get_code("0be2ebdc38492f25")
unflatten_mlp_params Ran spfrommer/latent_algebras/lib/inr2vec/utils.py
pointer only (licence: NONE) · get_code("70110f366c43d76e")
setup_trainer Not yet run spfrommer/latent_algebras/latalg/main/core.py
pointer only (licence: NONE) · get_code("c25a5205364024bd")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Machine learning often aims to produce latent embeddings of inputs which lie in a larger, abstract mathematical space. For example, in the field of 3D modeling, subsets of Euclidean space can be embedded as vectors using implicit neural representations. Such subsets also have a natural algebraic structure including operations (e.g., union) and corresponding laws (e.g., associativity). How can we learn to "union" two sets using only their latent embeddings while respecting associativity? We propose a general procedure for parameterizing latent space operations that are provably consistent with the laws on the input space. This is achieved by learning a bijection from the latent space to a carefully designed mirrored algebra which is constructed on Euclidean space in accordance with desired laws. We evaluate these structural transport nets for a range of mirrored algebras against baselines that operate directly on the latent space. Our experiments provide strong evidence that respecting the underlying algebraic structure of the input space is key for learning accurate and self-consistent operations.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2405.16763")
get_code_for_paper("2405.16763")
have("2405.16763")

Connect an agent — have() is free.