BERT4Rec

Reported on 4 benchmarks across 1 task · 1 paper · 1 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Knowledge Base4 results

Recommendation SystemsonMovieLens 1M
HR@10 (full corpus)· 2019-04-14
0.2843
best: 0.3412 (HSTU+MoL)
SOTA
BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformer arXiv:1904.06690
Recommendation SystemsonMovieLens 20M
HR@10 (full corpus)· 2019-04-14
0.2816
best: 0.3556 (HSTU)
BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformer arXiv:1904.06690
Recommendation SystemsonMovieLens 20M
nDCG@10 (full corpus)· 2019-04-14
0.1703
best: 0.2098 (HSTU)
BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformer arXiv:1904.06690
Recommendation SystemsonMovieLens 1M
NDCG@10 (full corpus)· 2019-04-14
0.1537
best: 0.1979 (HSTU+MoL)
BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformer arXiv:1904.06690