Diagnostic Classifiers Revealing how Neural Networks Process Hierarchical Structure

By Sara Veldhoen, Dieuwke Hupkes, Willem H. Zuidema. 2016

47 citations recorded by the source corpus.

Subfields: Natural Language Processing, Safety, Robustness & Interpretability

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

In Safety, Robustness & Interpretability

Citations in the corpus

Analysis Methods in Neural Language Processing: A Survey (2018, methodology): “5A similar method has been used to analyze hierarchical structure in neural networks trained on arithmetic expressions (Veldhoen et al., 2016; Hupkes et al., 2018).”
BERTology Meets Biology: Interpreting Attention in Protein Language Models (2020, methodology): “So-called diagnostic classiﬁers are used to interpret the outputs from BERT’s layers [81].”
Predicting Inductive Biases of Pre-Trained Models (2021, background): “Signiﬁcant prior work analyzes the representations and behavior of pre-trained LMs. Work using probing classiﬁers (Veldhoen et al., 2016; Adi et al., 2017; Conneau et al., 2018; Hupkes et al., 2018) suggests that such models capture a wide range of relevant linguistic phenomena (Hewitt & Man-ning,…”
Diagnostic classification and symbolic guidance to understand and improve recurrent neural networks (2017, methodology): “We work with the toy task of processing sentences from a lexical arithmetic language, consisting of spelled out, nested, arithmetic expressions [for more details, see Veldhoen et al., 2016].”
On internal language representations in deep learning: an analysis of machine translation and speech recognition (2018, methodology): “For instance, constructing small-scale models trained on synthetic datasets can lead to a more complete analysis [155, 328].”
LINSPECTOR WEB: A Multilingual Probing Suite for Word Representations (2019, background): “To address some of these challenges, Shi et al. (2016); Adi et al. (2017); Veldhoen et al. (2016); Conneau et al. (2018) have introduced probing tasks , a.k.a. diagnostic classiﬁers , that take as input a representation generated by a fully trained neural model and output predictions for a…”
Compositional Generalization by Learning Analytical Expressions (2020, unknown): “[37] explored the algebraic compositionality of neural networks via simple arithmetic expressions, and Saxton et al.”
What Happens To BERT Embeddings During Fine-tuning? (2020, methodology): “Voita et al. (2019a) used a form of canonical correlation analysis (PWCCA; Morcos et al.”
Edinburgh Research Explorer A systematic comparison of methods for low-resource dependency parsing on genuinely low-resource languages (unknown): “We investigate the extent to which data augmentation is useful for learning POS features, using diagnostic classifiers (Veldhoen et al., 2016; Adi et al., 2016; Shi et al., 2016) Table 4 reports the POS prediction accuracy.”
Frequency Effects on Syntactic Rule Learning in Transformers (2021, methodology): “To test whether the above predicted pattern holds, we use two probing classifiers (see Veldhoen et al., 2016; Ettinger et al., 2016, on probing) which we describe below.”