Unsupervised detection of contextualized embedding bias with application to ideology
We propose a fully unsupervised method to detect bias in contextualized embeddings. The method leverages the assortative information latently encoded by social networks and combines orthogonality regularization, structured sparsity learning, and graph neural networks to find the embedding subspace c...
| Huvudupphovsmän: | , , |
|---|---|
| Materialtyp: | Conference item |
| Språk: | English |
| Publicerad: |
Journal of Machine Learning Research
2022
|