Error Discovery by Clustering Influence Embeddings

Autor:	Wang, Fulton, Adebayo, Julius, Tan, Sarah, Garcia-Olano, Diego, Kokhlikyan, Narine
Rok vydání:	2023
Předmět:	Computer Science - Machine Learning
Druh dokumentu:	Working Paper
Popis:	We present a method for identifying groups of test examples -- slices -- on which a model under-performs, a task now known as slice discovery. We formalize coherence -- a requirement that erroneous predictions, within a slice, should be wrong for the same reason -- as a key property that any slice discovery method should satisfy. We then use influence functions to derive a new slice discovery method, InfEmbed, which satisfies coherence by returning slices whose examples are influenced similarly by the training data. InfEmbed is simple, and consists of applying K-Means clustering to a novel representation we deem influence embeddings. We show InfEmbed outperforms current state-of-the-art methods on 2 benchmarks, and is effective for model debugging across several case studies. Comment: NeuRIPs 2023 conference paper
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2312.04712 Zobrazit plný text záznamu View this record from Arxiv