showSidebars ==
showTitleBreadcrumbs == 1
node.field_disable_title_breadcrumbs.value ==
Thesis Proposal by WANG Zilin
| |
| | Robust Learning With Certification and Probabilistic Relaxation |  | WANG Zilin MPhil (IS) Student
School of Computing and Information Systems
Singapore Management University
| Research Area Dissertation Committee Research Advisor Committee Members |
| | Date 17 October 2024 (Thursday) | Time 4:00pm - 5:00pm | Venue Meeting room 4.4, Level 4. School of Computing and Information Systems 1, Singapore Management University,
80 Stamford Road
Singapore 178902 | Please register by 16 October 2024.
We look forward to seeing you at this research seminar. 
|
|
|
| | About The Talk Machine learning evaluates the performance by minimizing a loss function. However, robustness issues have gained great concern which can be fatal in safety-critical applications. Adversarial training can mitigate the issue by minimizing the loss of worst-case perturbations of data. It is effective in improving the robustness of the model, but it is too conservative that the plain performance of the model can be unsatisfying. This work empirically balances the average- and worst-case performance while the robustness of the model is not provable. A novel method is proposed for robust learning by sampling based on hypothesis testing. It guides the training to improve robustness in a probabilistic robustness setting efficiently. | | | About The Speaker WANG Zilin is a master student supervised by Professor SUN Jun. He is passionate about research on AI system security. |
|
|
