KDDCup 2020 - Track 2 - Adversarial Attacks and Defense on Academic Graph
https://www.biendata.xyz/competition/kddcup_2020_formal/
The challenge: Creating an attacker (i.e., a modified input consisting of graph structure (adjacency matrix) and node features (embedding vectors)) and a defender (i.e., a robust Graph Neural Network model). The organizers will match all attackers and defenders from all teams and rank the final leaderboard.
Final result: 7th
REFERENCES
Some Relevant Papers/Articles
Other Resources
- https://github.com/yenchenlin/awesome-adversarial-machine-learning
- https://stellargraph.readthedocs.io/en/stable/demos/index.html#find-algorithms-for-a-task
- https://adversarial-ml-tutorial.org/
- CS224W: Machine Learning with Graphs (Stanford - Fall 2019) [Course] [Videos]

