从神经表示中擦除未对齐属性
Erasure of Unaligned Attributes from Neural Representations
AI总结:
本文提出AMSAL算法,通过交替进行分配最大化和联合潜在空间投影,在属性未与输入样本直接对齐的隐式场景下从神经表示中擦除偏见信息,并在多个偏见基准上验证了其有效性。
摘要:
我们提出了分配最大化谱属性移除(AMSAL)算法,该算法在待擦除信息是隐式的、而非直接与每个输入样本对齐的情况下,从神经表示中擦除信息。我们的算法通过交替执行两个步骤来工作。在其中一个步骤中,它为输入表示找到与待擦除信息的分配关系;在另一个步骤中,它创建输入表示和待擦除信息到联合潜在空间的投影。我们在广泛的数据集阵列上测试了我们的算法,包括一个包含多个受保护属性的Twitter数据集、BiasBios数据集和BiasBench基准。最后一个基准包括四个具有各种类型受保护属性的数据集。我们的结果表明,在我们的设置中,偏见通常可以被移除。我们还讨论了当主任务与待擦除信息之间存在强纠缠时我们方法的局限性。
英文摘要:
We present the Assignment-Maximization Spectral Attribute removaL (AMSAL) algorithm, which erases information from neural representations when the information to be erased is implicit rather than directly being aligned to each input example. Our algorithm works by alternating between two steps. In one, it finds an assignment of the input representations to the information to be erased, and in the other, it creates projections of both the input representations and the information to be erased into a joint latent space. We test our algorithm on an extensive array of datasets, including a Twitter dataset with multiple guarded attributes, the BiasBios dataset and the BiasBench benchmark. The last benchmark includes four datasets with various types of protected attributes. Our results demonstrate that bias can often be removed in our setup. We also discuss the limitations of our approach when there is a strong entanglement between the main task and the information to be erased.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。