SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2106.16091 · NeurIPS · 2022

Exploring the Latent Space of Autoencoders with Interventional Assays

Bernhard Schölkopf, Felix Leeb, Michel Besserve, Stefan Bauer

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 5 functions out of this paper's own repositories and ran 3 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
felixludos/latent-responses canonical 3 of 5
FunctionStatusWhere it lives
response_mat Ran felixludos/latent-responses/src/responses.py
code served (permissive licence) · get_code("6714b05c725caffb")
response_mat Ran felixludos/latent-responses/src/responses.py
code served (permissive licence) · get_code("070d219fb032a1fd")
score_from_conditioned_response_mat Ran felixludos/latent-responses/src/responses.py
code served (permissive licence) · get_code("8210347260100074")
Autoencoder Not yet run felixludos/latent-responses/src/responses.py
code served (permissive licence) · get_code("39767c58bcd52855")
conditioned_response_mat Not yet run felixludos/latent-responses/src/responses.py
code served (permissive licence) · get_code("70adfb01098fa961")

Repositories linked to this paper

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

Abstract

Autoencoders exhibit impressive abilities to embed the data manifold into a lowdimensional latent space, making them a staple of representation learning methods. However, without explicit supervision, which is often unavailable, the representation is usually uninterpretable, making analysis and principled progress challenging. We propose a framework, called latent responses, which exploits the locally contractive behavior exhibited by variational autoencoders to explore the learned manifold. More specifically, we develop tools to probe the representation using interventions in the latent space to quantify the relationships between latent variables. We extend the notion of disentanglement to take the learned generative process into account and consequently avoid the limitations of existing metrics that may rely on spurious correlations. Our analyses underscore the importance of studying the causal structure of the representation to improve performance on downstream tasks such as generation, interpolation, and inference of the factors of variation.

For agents

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

get_harvested_code_for_paper("2106.16091")
get_code_for_paper("2106.16091")
have("2106.16091")

Connect an agent — have() is free.