2 papers
cs.LG2024
UnLearning from Experience to Avoid Spurious Correlations
Jeff Mitchell, Jesús MartÃnez del Rincón, Niall McLaughlin
While deep neural networks can achieve state-of-the-art performance in many tasks, these models are more fragile than they appear. They are prone to learning spurious correlations…
cs.LG2024
Hard-label based Small Query Black-box Adversarial Attack
Jeonghwan Park, Paul Miller, Niall McLaughlin
We consider the hard label based black box adversarial attack setting which solely observes predicted classes from the target model. Most of the attack methods in this setting suff…