Click Through Rate Prediction On Criteo
評価指標
AUC
Log Loss
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
比較表
| モデル名 | AUC | Log Loss |
|---|---|---|
| product-based-neural-networks-for-user | 0.7972 | 0.45323 |
| optimizing-feature-set-for-click-through-rate | 0.8116 | 0.4401 |
| dcn-m-improved-deep-cross-network-for-feature | 0.8115 | 0.4406 |
| fibinet-improving-fibinet-by-greatly-reducing | 0.8110 | - |
| towards-deeper-lighter-and-interpretable-1 | 0.8161 | 0.4360 |
| xdeepfm-combining-explicit-and-implicit | 0.8052 | 0.4418 |
| tfnet-multi-semantic-feature-interaction-for | 0.7991 | - |
| cetn-contrast-enhanced-through-network-for | 0.8148 | 0.4373 |
| deep-learning-over-multi-field-categorical | 0.7963 | 0.45738 |
| fibinet-combining-feature-importance-and | 0.8103 | 0.4423 |
| memonet-memorizing-representations-of-all | 0.8152 | - |
| autofis-automatic-feature-interaction | 0.8010 | 0.5405 |
| wide-deep-learning-for-recommender-systems | 0.7981 | 0.46772 |
| fi-gnn-modeling-feature-interactions-via | 0.8062 | 0.4453 |
| deepfm-a-factorization-machine-based-neural | 0.8007 | 0.45083 |
| a-sparse-deep-factorization-machine-for | 0.8123 | 0.4395 |
| memorize-factorize-or-be-naive-learning | 0.8101 | 0.4417 |
| xcrossnet-feature-structure-oriented-learning | 0.8067 | - |
| tf4ctr-twin-focus-framework-for-ctr | 0.8150 | - |
| correct-normalization-matters-understanding | 0.8107 | - |
| mmbattn-max-mean-and-bit-wise-attention-for | 0.81497 | - |
| finalmlp-an-enhanced-two-stream-mlp-model-for-1 | 0.8149 | - |
| weighted-multi-level-feature-factorization | 0.804 | 0.447 |
| adaptive-factorization-network-learning | 0.8074 | - |
| dcnv3-towards-next-generation-deep-cross | 0.8162 | 0.4358 |
| mmbattn-max-mean-and-bit-wise-attention-for | 0.8143 | - |
| contextnet-a-click-through-rate-prediction | 0.8113 | - |
| gatenet-gating-enhanced-deep-network-for | 0.8100 | - |
| product-based-neural-networks-for-user | 0.7987 | 0.45214 |
| fat-deepffm-field-attentive-deep-field-aware | 0.8104 | 0.4416 |
| autoint-automatic-feature-interaction | 0.8061 | 0.4454 |
| clustering-embedding-tables-without-first | 0.806 | 0.449 |
| stec-see-through-transformer-based-encoder | 0.8143 | 0.4379 |
| feature-interaction-based-neural-network-for | 0.8020 | 0.5409 |
| optembed-learning-optimal-embedding-table-for | 0.8114 | 0.44 |
| product-based-neural-networks-for-user | 0.7982 | 0.45256 |
| cognitive-evolutionary-search-to-select | 0.8117 | 0.4400 |
| masknet-introducing-feature-wise | 0.8131 | - |