Evaluation code/details missing for CLBD

aks2203 / poisoning-benchmark

A unified benchmark problem for data poisoning attacks

https://arxiv.org/abs/2006.12557

MIT License

146 stars 21 forks source link

Evaluation code/details missing for CLBD #4

Closed dhruva77 closed 3 years ago

dhruva77 commented 3 years ago

The files poison_test.py/benchmark_test.py contains evaluation code for only targeted attacks. For evaluation of untargeted attacks, I couldn't find any code in the repo or details in the paper. Is the evaluation methodology for target and untargeted attacks the same or do you use the evaluation methodology mentioned in CLBD paper to evaluate untargeted attacks? Can you please clarify?

Thanks!

aks2203 commented 3 years ago

Hi, this repo is only concerned with targeted attacks.

dhruva77 commented 3 years ago

Hi,

To be clear, was CLBD evaluated as a targeted attack in the results reported in the paper and repo?

aks2203 commented 3 years ago

Yes. All attacks, including CLBD, are evaluated as targeted attacks.