Attention-based Iterative Decomposition for Tensor Product Representation

Source code for the ICLR'24 paper: Attention-based Iterative Decomposition for Tensor Product Representation

TL;DR: Slot-based competitive mechanism that effectively binds sequential features to the structured representations (roles and fillers) of TPR

Requirement

• Python 3.8
• PyTorch 1.13

pip install -r requirements.txt

Acknowledement

We've developed our code using open-source implementations for various tasks, as below.

• bAbI task (FWM)

• bAbI task (TPR-RNN)

• Sort-of-CLEVR task

• WikiText-103 task

Citation

@inproceedings{
  park2024attentionbased,
  title={Attention-based Iterative Decomposition for Tensor Product Representation},
  author={Taewon Park and Inchul Choi and Minho Lee},
  booktitle={The Twelfth International Conference on Learning Representations},
  year={2024},
  url={https://openreview.net/forum?id=FDb2JQZsFH}
}