Input Switched Affine Networks: An RNN Architecture Designed for Interpretability

By Jakob N. Foerster, Justin Gilmer, Jascha Narain Sohl-Dickstein, Jan Chorowski, David Sussillo. 2016

38 citations recorded by the source corpus.

Subfields: Natural Language Processing, Model Architectures

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

1. Colin Raffel
2. Yejin Choi
3. Adam Roberts
4. Mohit Bansal
5. Sharan Narang
6. Noah Fiedel
7. Katherine Lee
8. Hannaneh Hajishirzi

In Model Architectures

1. Nino Antulov-Fantulin
2. Adam Roberts
3. Tao Lin
4. Karishma Malkan
5. Tian Guo
6. Noah Fiedel
7. Colin Raffel
8. Katherine Lee

Citations in the corpus

• WT5?! Training Text-to-Text Models to Explain their Predictions (2020, background): “C L ] 3 0 A pr 2 02 0 Smilkov et al., 2017) or by designing architectures that are easier to analyze (Foerster et al., 2017; Jacobsen et al., 2018; Bahdanau et al., 2014; Raffel et al., 2017).”
• Exploring Interpretable LSTM Neural Networks over Multi-Variable Data (2019, background): “In addition to forecasting, interpretable RNNs are desirable for gaining insights into the important part of data for RNNs achieving good prediction performance (Hu et al., 2018; Foerster et al., 2017; Lipton, 2016).”
• Online Learning with Gated Linear Networks (2017, methodology): “Foerster et al. (2017) recently introduced a novel recurrent neural architecture whose modeling power was derived from using a data-dependent affine transformation as opposed to a non-linear activation function.”