AI Paper Atlas

Research map / Optimization / Implicit bias and generalization

Stochastic Shared Embeddings: Data-driven Regularization of Embedding Layers

Liwei Wu, Shuqing Li, Cho-Jui Hsieh, James L Sharpnack

NeurIPS 2019 · Optimization › Implicit bias and generalization · approach: Shallow ReLU network convergence · 19 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 SSE-Graph instead of graph Laplacian regularizationMF on Movielens1m · Movielens1mRMSE: 1.0464 -> 1.0145Improved
Use SSE-Graph instead of graph Laplacian regularizationMF on Movielens10m · Movielens10mRMSE: 1.9755 -> 1.9019Improved
Add SSE-SE to MFMF · Douban; Movielens10m; NetflixRMSE: 0.7339 -> 0.7201; 0.8851 -> 0.8715; 0.8941 -> 0.8842Improved
Add SSE-SE alongside dropoutMF · Douban; Movielens10m; NetflixRMSE: 0.7296 -> 0.7185; 0.8813 -> 0.8678; 0.8897 -> 0.8790Improved
Replace dropout with SSE-SEBPR · Movielens1m; Yahoo Music; FoursquareP@5: 0.6548 -> 0.6813; 0.3315 -> 0.3498; 0.0184 -> 0.0262Mixed
Add SSE-SE to SASRec input embeddings only (ablation)SASRec (2 blocks) · Movielens1mNDCG@10: 0.5941 -> 0.6092Improved
Add SSE-SE to SASRec output embeddings only (ablation)SASRec (2 blocks) · Movielens1mNDCG@10: 0.5941 -> 0.6085Improved
Add SSE-SE to both SASRec input and output embeddings (ablation)SASRec (2 blocks) · Movielens1mNDCG@10: 0.5941 -> 0.6200Improved
Increase SASRec depth from 2 to 6 blocks while using SSE-SE on both embedding types (ablation)SASRec · Movielens1mNDCG@10: 0.6200 -> 0.6265Improved
Add SSE-SE to Transformer encoder and decoder embeddingsTransformer · WMT English-to-German newstest, 2008–2018BLEU: 21.0 -> 21.4; 20.7 -> 21.1; 22.7 -> 23.0; 20.6 -> 21.0; 20.6 -> 20.8; 25.3 -> 25.2; 26.2 -> 27.2; 28.4 -> 29.2; 32.1 -> Mixed
Add SSE-SE during BERT pre-trainingBERT-Base · IMDB Test SetAccuracy: 0.9518 -> 0.9542Improved
Add SSE-SE during BERT pre-trainingBERT-Base · SST-2 Test SetAccuracy (%): 93.8 -> 94.3Improved

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