Title | ||
---|---|---|
Evaluating neural network explanation methods using hybrid documents and morphological prediction. |
Abstract | ||
---|---|---|
We propose two novel paradigms for evaluating neural network explanations in NLP. The first paradigm works on hybrid documents, the second exploits morphosyntactic agreements. Neither paradigm requires manual annotations; instead, a relevance ground truth is generated automatically. In our experiments, successful explanations for Long Short Term Memory networks (LSTMs) were produced by a decomposition of memory cells (Murdoch u0026 Szlam, 2017), while for convolutional neural networks, a gradient-based method by (Denil et al., 2014) works well. We also introduce LIMSSE, a substring-based extension of LIME (Ribeiro et al., 2016) that produces the most successful explanations in the hybrid document experiment. |
Year | Venue | Field |
---|---|---|
2018 | arXiv: Computation and Language | Substring,Convolutional neural network,Computer science,Long short term memory,Exploit,Ground truth,Artificial intelligence,Artificial neural network,Machine learning |
DocType | Volume | Citations |
Journal | abs/1801.06422 | 0 |
PageRank | References | Authors |
0.34 | 23 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Nina Pörner | 1 | 11 | 4.34 |
Hinrich Schütze | 2 | 2113 | 362.21 |
Benjamin Roth | 3 | 307 | 20.45 |