Interpretability

On the Value of Out-of-Distribution Testing: An Example of Goodhart's Law NeurIPS 2020 On the weak link between importance and prunability of attention heads EMNLP 2020 Opening the Black Box: Automatically Characterizing Software for Algorithm Selection (Student Abstract) AAAI 2020 Optimization Methods for Interpretable Differentiable Decision Trees Applied to Reinforcement Learning AISTATS 2020 PAC Confidence Sets for Deep Neural Networks via Calibrated Prediction ICLR 2020 Parallels Between Phase Transitions and Circuit Complexity? COLT 2020 Parameterized Explainer for Graph Neural Network NeurIPS 2020 Pareto Probing: Trading Off Accuracy for Complexity EMNLP 2020 Parsers Know Best: German PP Attachment Revisited COLING 2020 Parsing into Variable-in-situ Logico-Semantic Graphs ACL 2020 PathologyGAN: Learning deep representations of cancer tissue MIDL 2020 Perturbed Masking: Parameter-free Probing for Analyzing and Interpreting BERT ACL 2020 PGM-Explainer: Probabilistic Graphical Model Explanations for Graph Neural Networks NeurIPS 2020 Phenotyping with Prior Knowledge using Patient Similarity MLHC 2020 Precise Task Formalization Matters in Winograd Schema Evaluations EMNLP 2020 Predicting the Focus of Negation: Model and Error Analysis ACL 2020 Priorless Recurrent Networks Learn Curiously COLING 2020 Probing Multilingual BERT for Genetic and Typological Signals COLING 2020 Probing Natural Language Inference Models through Semantic Fragments AAAI 2020 Probing Neural Dialog Models for Conversational Understanding ACL 2020 Probing Task-Oriented Dialogue Representation from Language Models EMNLP 2020 Problems with Shapley-value-based explanations as feature importance measures ICML 2020 Proper Network Interpretability Helps Adversarial Robustness in Classification ICML 2020 ProtoAttend: Attention-Based Prototypical Learning JMLR 2020 Purifying Interaction Effects with the Functional ANOVA: An Efficient Algorithm for Recovering Identifiable Additive Models AISTATS 2020