BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformer

TL;DR

BERT4Rec introduces a bidirectional transformer-based model trained with a Cloze task to better capture users' evolving preferences for recommendation systems, outperforming previous unidirectional models.

Contribution

The paper presents BERT4Rec, a novel bidirectional transformer model for sequential recommendation, trained with a Cloze task to utilize both past and future context.

Findings

BERT4Rec outperforms state-of-the-art models on four benchmark datasets.

Bidirectional modeling improves recommendation accuracy.

Training with the Cloze task enhances the model's ability to learn from both directions.

Abstract

Modeling users' dynamic and evolving preferences from their historical behaviors is challenging and crucial for recommendation systems. Previous methods employ sequential neural networks (e.g., Recurrent Neural Network) to encode users' historical interactions from left to right into hidden representations for making recommendations. Although these methods achieve satisfactory results, they often assume a rigidly ordered sequence which is not always practical. We argue that such left-to-right unidirectional architectures restrict the power of the historical sequence representations. For this purpose, we introduce a Bidirectional Encoder Representations from Transformers for sequential Recommendation (BERT4Rec). However, jointly conditioning on both left and right context in deep bidirectional model would make the training become trivial since each item can indirectly "see the target item". To address this problem, we train the bidirectional model using the Cloze task, predicting the masked items in the sequence by jointly conditioning on their left and right context. Comparing with predicting the next item at each position in a sequence, the Cloze task can produce more samples to train a more powerful bidirectional model. Extensive experiments on four benchmark datasets show that our model outperforms various state-of-the-art sequential models consistently.