https://github.com/qkrdmsghk/goodhse

[CVPR 2024] Improving out-of-distribution generalization in graphs via hierarchical semantic environments
https://github.com/qkrdmsghk/goodhse

graph molecule ood-generalization

Last synced: 2 months ago
JSON representation

[CVPR 2024] Improving out-of-distribution generalization in graphs via hierarchical semantic environments

• Host: GitHub
• URL: https://github.com/qkrdmsghk/goodhse
• Owner: qkrdmsghk
• Created: 2024-05-16T09:48:48.000Z (7 months ago)
• Default Branch: main
• Last Pushed: 2024-07-29T05:58:38.000Z (5 months ago)
• Last Synced: 2024-10-09T22:05:38.200Z (2 months ago)
• Topics: graph, molecule, ood-generalization
• Language: Python
• Homepage:
• Size: 913 KB
• Stars: 5
• Watchers: 2
• Forks: 3
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        

Improving out-of-distribution generalization in graphs via hierarchical semantic environments



    

    

    

      

    

    

    

     



## Motivation

The complex nature of molecule graphs poses unique challenges to out-of-distribution (OOD) generalization, differentiating them from images and general graphs:

1. Molecules with Similar Structures Can Have Different Functions:

- Functional Group: Acts as a causal subgraph.

- Scaffold: Serves as an environmental (noise) subgraph.

2. Semantic Environments in OOD:

- Environments are related.

- Environments are hierarchical.



Figure 1. Challenges on molecule graphs in out-of-distribution setting.


## Challenges

Existing methods use flat enviornments to conduct the graph invariant learning.

There are two limitations in flat environment infernece:

1. Provided (Real): neglect local similarity among the numerous environments.

2. Inference (Infer #2): Inferring from a small number of environments may fail to capture global similarity and interrelationships among the environments. 



Figure 2. (a) Results on IC50-SCA dataset from DrugOOD. (b) Flat environments from existing approaches. (c) Hierarchical environments from our methods. For visualization, we set #real environments as 10.


## Instructions



Figure 3.Our Framework consists of (a) Hierarchical Stochastic Subgraph Generation, (b) Hierarchical Semantic Environments, (c) Robust GIL with Hierarchical Semantic Environments.


### Installation and data preparation

Our code is based on the following libraries:

```

torch==1.9.0+cu111

torch-geometric==2.0.2

```

plus the [DrugOOD](https://github.com/tencent-ailab/DrugOOD) benchmark repo.

The data used in the paper can be obtained following these [instructions](./dataset_gen/README.md).

### Reproduce results

We provide the hyperparamter tuning and evaluation details in the paper and appendix.

In the below we give a brief introduction of the commands and their usage in our code. 

We provide the corresponding running scripts in the [script](./scripts/) folder.

Simply run 

```

bash run.sh 0 icassay

```

with corresponding datasets and model specifications.

If you find our paper and repo useful, please cite our paper: -->

```bibtex -->

@inproceedings{piao2024improving,

  title={Improving out-of-distribution generalization in graphs via hierarchical semantic environments},

  author={Piao, Yinhua and Lee, Sangseon and Lu, Yijingxiu and Kim, Sun},

  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition},

  pages={27631--27640},

  year={2024}

}

```

Ack: The readme is inspired by [CIGA](https://github.com/LFhase/CIGA). 😄