This is pytorch implementation of "Differentiable Neural Computers" with Memory Demon presented at NeurIPS 2022 Workshop On Memory in Artificial and Real Intelligence.
Go to file
Ari Azarafrooz 41c2371849 fixed readme
2022-11-05 15:02:07 -07:00
assets dnc-demon impl 2022-11-05 14:59:40 -07:00
Dataset dnc-demon impl 2022-11-05 14:59:40 -07:00
Models dnc-demon impl 2022-11-05 14:59:40 -07:00
Utils dnc-demon impl 2022-11-05 14:59:40 -07:00
Visualize dnc-demon impl 2022-11-05 14:59:40 -07:00
memory_demon.py dnc-demon impl 2022-11-05 14:59:40 -07:00
README.md fixed readme 2022-11-05 15:02:07 -07:00
requirements.txt dnc-demon impl 2022-11-05 14:59:40 -07:00

Differentiable Neural Computers with Memory Demon

alt text

This is pytorch implementation of "Differentiable Neural Computers" with Memory Demon presented at NeurIPS 2022 Workshop On Memory in Artificial and Real Intelligence.

To repeat the babi experiments:

python memory_demon.py -name tmp_train_dir -profile babi,dnc-msd

Acknowledgment:

This code uses the pytorch DNC implementation of https://github.com/RobertCsordas/dnc