Erasure of Unaligned Attributes from Neural Representations
Shao, Shun, Ziser, Yftah, Cohen, Shay
–arXiv.org Artificial Intelligence
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.
arXiv.org Artificial Intelligence
Apr-18-2023
- Country:
- Oceania > Australia
- North America
- United States
- Texas (0.04)
- Washington > King County
- Seattle (0.04)
- New Mexico > Santa Fe County
- Santa Fe (0.04)
- Minnesota > Hennepin County
- Minneapolis (0.14)
- Maryland > Prince George's County
- College Park (0.04)
- Louisiana > Orleans Parish
- New Orleans (0.04)
- California > Santa Clara County
- Palo Alto (0.04)
- Puerto Rico > San Juan
- San Juan (0.04)
- United States
- Europe
- United Kingdom (0.04)
- Spain > Catalonia
- Barcelona Province > Barcelona (0.04)
- Ireland > Leinster
- County Dublin > Dublin (0.04)
- Denmark > Capital Region
- Copenhagen (0.04)
- Belgium > Brussels-Capital Region
- Brussels (0.04)
- Genre:
- Research Report > New Finding (1.00)
- Industry:
- Information Technology (0.46)
- Technology: