AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Improving Language Modelling with Noise Contrastive Estimation

Farhana Ferdousi Liza, Marek Grzes

AAAI 2018 · Label-efficient and robust learning › Recognition and long-tail · approach: Multilingual language representation pretraining · 4 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 NCE instead of softmax for a large LSTM trained for 55 epochsLarge PTB LSTM · PTB testperplexity: 78.826 -> 69.995Improved
Use NCE instead of softmax for a large LSTM trained for 20 epochsLarge PTB LSTM · PTB testperplexity: 76.935 -> 74.129Improved
Use a long search period, τ=25, rather than τ=6 (ablation)NCE LSTM · PTB test, medium modelperplexity: 83.858 -> 75.959Improved
Use a long search period, τ=25, rather than τ=6 (ablation)Softmax LSTM · PTB test, medium modelperplexity: 78.567 -> 79.906Hurt
Add smaller-than-standard uniform weight initialization (ablation)NCE LSTM · PTB test, large modelperplexity: 71.184 -> 70.444 -> 69.995Improved
Add empirically tuned smaller uniform weight initialization (ablation)NCE LSTM · PTB test, medium modelperplexity: 75.960 -> 75.286Improved
Reduce the standard uniform initialization range by a factor of 4 (ablation)NCE LSTM · PTB test, small modelperplexity: 104.449 -> 102.245Improved
Use empirically tuned smaller uniform initialization rather than the range reduced by 4 (ablation)NCE LSTM · PTB test, small modelperplexity: 102.245 -> 102.237Improved
Increase dropout from 60% to 70% (ablation)Softmax large LSTM · PTB validationperplexityMixed
Increase the softmax search-period learning rate to raise early perplexity (ablation)Softmax large LSTM · PTB validationperplexityNo clear effect
Use 600 noise samples rather than other tested sample counts (ablation)NCE LSTM · PTBperplexityMixed
Use lower-variance initialization for NCE (ablation)NCE LSTM · PTB test, large modelperplexity: 71.184 -> 69.995Improved

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