This is the official source code for our AAAI 2024 Paper "An Attentive Inductive Bias for Sequential Recommendation beyond the Self-Attention"

Beyond Self-Attention for Sequential Recommendation (BSARec) leverages Fourier transform to strike a balance between our inductive bias and self-attention.

• (Nov 12, 2024) add visulaization code for figure 2(b)
• (Oct 18, 2024) correct default argument for layer-specific values in FEARec
• (Oct 18, 2024) rename variables in model checkpoint
• (Oct 18, 2024) organize layer classes within each model file
• (Sep 14, 2024) add data processing code
• (Apr 20, 2024) rename variable 'beta' to 'sqrt_beta'
• (Apr 16, 2024) add visualization code for figure 3

In our experiments, we utilize six datasets, all stored in the src/data folder.

• For the Beauty, Sports, Toys, and Yelp datasets, we employed the datasets downloaded from this repository.
• For ML-1M and LastFM, we processed the data according to the procedure outlined in this code.
• The src/data/*_same_target.npy files are utilized for training DuoRec and FEARec, both of which incorporate contrastive learning.

conda env create -f bsarec_env.yaml
conda activate bsarec

• Note that pretrained model (.pt) and train log file (.log) will saved in src/output
• train_name: name for log file and checkpoint file

python main.py  --data_name [DATASET] \
                --lr [LEARNING_RATE] \
                --alpha [ALPHA] \ 
                --c [C] \
                --num_attention_heads [N_HEADS] \
                --train_name [LOG_NAME]

• Example for Beauty

python main.py  --data_name Beauty \
                --lr 0.0005 \
                --alpha 0.7 \
                --c 5 \
                --num_attention_heads 1 \
                --train_name BSARec_Beauty

• Note that pretrained model (.pt file) must be in src/output
• load_model: pretrained model name without .pt

python main.py  --data_name [DATASET] \
                --alpha [ALPHA] \ 
                --c [C] \
                --num_attention_heads [N_HEADS] \
                --load_model [PRETRAINED_MODEL_NAME] \
                --do_eval

• Example for Beauty

python main.py  --data_name Beauty \
                --alpha 0.7 \
                --c 5 \
                --num_attention_heads 1 \
                --load_model BSARec_Beauty_best \
                --do_eval

• You can easily train the baseline models used in BSARec by changing the model_type argument.
  • model_type: Caser, GRU4Rec, SASRec, BERT4Rec, FMLPRec, DuoRec, FEARec
• For the hyperparameters for the baselines, check the parse_args() function in src/utils.py.

python main.py  --model_type SASRec \
                --data_name Beauty \
                --num_attention_heads 1 \
                --train_name SASRec_Beauty

If you find our work useful, please consider citing our paper:

@inproceedings{shin2024bsarec,
title={An attentive inductive bias for sequential recommendation beyond the self-attention},
author={Shin, Yehjin and Choi, Jeongwhan and Wi, Hyowon and Park, Noseong},
booktitle={Proceedings of the AAAI Conference on Artificial Intelligence},
volume={38},
number={8},
pages={8984--8992},
year={2024}
}

If you have any inquiries regarding our paper or codes, feel free to reach out via email at yehjin.shin@kaist.ac.kr.

This repository is based on FMLP-Rec.