On the Modularity of Hypernetworks (arxiv).

Pytorch Implementation of the experiments in "On the Modularity of Hypernetworks" (NeurIPS 2020)

Prerequisites

• Python 3.6+
• Pytorch 0.4

Rotations Prediction

Run rotation_prediction/main.py. You can use the following example to run:

python main.py --lr 0.001 --epochs 10

Validating Assumption 1

Run validating_assumption_1/main.py. You can use the following example to run:

python validating_assumption_1/train.py --lr 0.01 --epochs 100 --dataset MNIST

Validating Assumption 2

Run validating_assumption_2/train.py. You can use the following example to run:

python validating_assumption_2/train.py --lr 1.0 --epochs 5 --dataset MNIST

Reference

If you found this code useful, please cite the following paper:

@inproceedings{galanti2020modularity,
  title={On the Modularity of Hypernetworks},
  author={Tomer Galanti and Lior Wolf},
  booktitle={NeurIPS},
  year={2020}
}