AdaptDel: Adaptable Deletion Rate Randomized Smoothing for Certified Robustness
Zhuoqun Huang, Neil G. Marchant, Olga Ohrimenko, Benjamin I. P. Rubinstein
我们考虑了基于编辑距离扰动的序列分类的认证稳健性问题。 自然出现的不同长度的输入(例如,自然语言处理任务中的句子)对采用固定率删除机制并导致性能欠优的当前方法提出了挑战。 为此,我们引入了具有可适应删除速率的AdaptDel方法,可以根据输入属性动态调整。 我们将随机平滑的理论框架扩展到可变速率删除,确保有关编辑距离的声音认证。 我们在自然语言任务中取得了强大的经验成果,与最先进的认证相比,观察到认证区域的中位基数提高了30个数量级。
We consider the problem of certified robustness for sequence classification against edit distance perturbations. Naturally occurring inputs of varying lengths (e.g., sentences in natural language processing tasks) present a challenge to current methods that employ fixed-rate deletion mechanisms and lead to suboptimal performance. To this end, we introduce AdaptDel methods with adaptable deletion rates that dynamically adjust based on input properties. We extend the theoretical framework of rando...