CTSMA-ICML21

Code for ICML21 paper "Learning Self-Modulating Attention in Continuous Time Space with Applications to Sequential Recommendation"

Installation

The program requires Python 3.7+ with NumPy, Pandas and Tensorflow 1.x.

Data Format

The implementation is desiged for top-N recommendations on implicit data, and thus it takes Tensorflow-Records as input:

seqs_i:   int64, the sequence of item ids
seqs_t: float32, the sequence of purchase timestamps
label:    int64, the output item ids

In addition to train/validation/test files, the mapping from item-id to mark-id should be also specified and stored as scipy-sparse matrix.

Train and Test

Once the data is ready, it is quite simple to train and evaluate our S2PNM model by running

bash runme.sh

Citation

If you find our code useful for your research, please consider cite.

@inproceedings{chen2021learning,
  title={Learning Self-Modulating Attention in Continuous Time Space with Applications to Sequential Recommendation},
  author={Chen, Chao and Geng, Haoyu and Yang, Nianzu and Yan, Junchi and Xue, Daiyue and Yu, Jianping and Yang, Xiaokang},
  booktitle={Proceedings of the International Conference on Machine Learning (ICML '21)},
  pages={1606--1616},
  year={2021},
  organization={PMLR}
}