giandos200 / zhang-et-al-mitigating-unwanted-biases-with-adversarial-learning

• 👋 Hi, I’m @giandos200
• 👀 I’m interested in Artificial Intelligence & Machine Learning
• 👦 I’m currently a Research Scientist IBM Research in Dublin, Ireland and Ph.D. candidate at the Polytechnic University of Bari.
• 💻 On the 8th of January, I joined IBM Research in Dublin, Ireland, as a Research Scientist in the Security & Privacy team.
• 📚 I daily work on Large Language Models (LLM) Security & Privacy. I’m looking to collaborate on Trustworthy AI (Fairness, Explainability, and Counterfactual Reasoning) wherever would be the domain of application.
• ✉️ Please, contact me on my mail: giando95menico[at]gmail[dot]com
• 📜 If you are curious about my published research papers, visit my Google scholar profile!

zhang-et-al-mitigating-unwanted-biases-with-adversarial-learning's Issues

Question about AdversarialDebiasing.py

Hello,
On the line 278 of the code for AdversarialDebiasing.py you have written
adv_grad = [torch.clone(par.grad.detach()) for par in self.clf_model.parameters()]
Shouldn't it be
adv_grad = [torch.clone(par.grad.detach()) for par in self.adv_model.parameters()]
instead ?