SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2403.01946 · NeurIPS · 2024

A Generative Model of Symmetry Transformations

Javier Antorán, James Allingham, José Hernández-Lobato, Richard Turner, Eric Nalisnick, David Krueger, Shreyas Padhy, Bruno Mlodozeniec

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 6 functions out of this paper's own repositories and ran 6 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
cambridge-mlg/sgm canonical 6 of 6
FunctionStatusWhere it lives
approximate_mode Ran cambridge-mlg/sgm/src/models/utils.py
code served (permissive licence) · get_code("f1d256bcc201a1b8")
clipped_adamw Ran cambridge-mlg/sgm/src/models/utils.py
code served (permissive licence) · get_code("cfdc596d5298708d")
format_thousand Ran cambridge-mlg/sgm/experiments/utils.py
code served (permissive licence) · get_code("ba79fb4fbf622e93")
gen_affine_matrix Ran cambridge-mlg/sgm/src/transformations/affine.py
code served (permissive licence) · get_code("efb8c38e3a6117e3")
gen_affine_matrix_no_shear Ran cambridge-mlg/sgm/src/transformations/affine.py
code served (permissive licence) · get_code("ad010a64b6302335")
reset_metrics Ran cambridge-mlg/sgm/src/models/utils.py
code served (permissive licence) · get_code("4223d8d7d93fed49")

Repositories linked to this paper

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

Abstract

Correctly capturing the symmetry transformations of data can lead to efficient models with strong generalization capabilities, though methods incorporating symmetries often require prior knowledge. While recent advancements have been made in learning those symmetries directly from the dataset, most of this work has focused on the discriminative setting. In this paper, we take inspiration from group theoretic ideas to construct a generative model that explicitly aims to capture the data's approximate symmetries. This results in a model that, given a prespecified but broad set of possible symmetries, learns to what extent, if at all, those symmetries are actually present. Our model can be seen as a generative process for data augmentation. We provide a simple algorithm for learning our generative model and empirically demonstrate its ability to capture symmetries under affine and color transformations, in an interpretable way. Combining our symmetry model with standard generative models results in higher marginal test-log-likelihoods and improved data efficiency.

For agents

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

get_harvested_code_for_paper("2403.01946")
get_code_for_paper("2403.01946")
have("2403.01946")

Connect an agent — have() is free.