AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Generalizing to Evolving Domains with Latent Structure-Aware Sequential Autoencoder

Tiexin Qin, Shiqi Wang, Haoliang Li

ICML 2022 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Optimization-based domain generalization · 5 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use LSSAE instead of DIVAevolving-domain generalization · Circleaccuracy (%): 67.9 -> 73.8Improved
use LSSAE instead of AndMaskevolving-domain generalization · Sineaccuracy (%): 69.3 -> 71.4Improved
use LSSAE instead of Mixupevolving-domain generalization · RMNISTaccuracy (%): 44.9 -> 46.4Improved
use LSSAE instead of MTLevolving-domain generalization · Portraitsaccuracy (%): 89.0 -> 89.1Improved
use LSSAE instead of DIVAevolving-domain generalization · Caltranaccuracy (%): 69.2 -> 70.6Improved
use LSSAE instead of IRMevolving-domain generalization · Circle-Caccuracy (%): 38.5 -> 44.8Improved
use LSSAE instead of Fishevolving-domain generalization · Sine-Caccuracy (%): 62.7 -> 60.8Hurt
use LSSAE instead of ERMevolving-domain generalization · PowerSupplyaccuracy (%): 71.0 -> 71.1No clear effect
add the category-related latent variable z^v (ablation)LSSAE · PowerSupplyaccuracy (%): 70.7 -> 71.1Improved
use a categorical rather than learnable Gaussian prior (ablation)LSSAE · PowerSupplyaccuracy (%): 70.1 -> 71.1Improved
use a categorical rather than fixed Uniform prior (ablation)LSSAE · PowerSupplyaccuracy (%): 71.0 -> 71.1No clear effect
add the temporal smooth constraint (ablation)LSSAE · RMNISTvariance of test accuracy: 3.6 -> 0.4Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy