Why I'm not Answering: Understanding Determinants of Classification of an Abstaining Classifier for Cancer Pathology Reports
Explore this paper's citation graph
Summary
This work identifies the determinants of abstention with LIME (the Local Interpretable Model-agnostic Explanations method), and trains a model to learn the attributes of pathology reports that are likely to lead to incorrect classifications, albeit at the cost of reduced sensitivity.
- Type
- preprint
- Published
- 2020-09-10
- Cited by
- 2
- References
- 37
- Access
- Open access
- OpenAlex
- https://openalex.org/W3087268877
- Semantic Scholar
- https://api.semanticscholar.org/CorpusID:221865825
Keywords
Classifier (UML), Computer science, Artificial intelligence, Machine learning, Medicine
References
- Efficient Estimation of Word Representations in Vector Space
- Convolutional Neural Networks for Sentence Classification
- Deep neural networks are easily fooled: High confidence predictions for unrecognizable images
- Return of the Devil in the Details: Delving Deep into Convolutional Nets
- Deep learning in neural networks: An overview
- Hybrid approach to robust dialog management using agenda and dialog examples
- New types of deep neural network learning for speech recognition and related applications: an overview
- Going deeper with convolutions
- A unified architecture for natural language processing: deep neural networks with multitask learning
- Distributed Representations of Words and Phrases and their Compositionality
- ImageNet classification with deep convolutional neural networks
- Representation Learning: A Review and New Perspectives
- Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps
- Hidden Technical Debt in Machine Learning Systems
- Towards Open Set Deep Networks
- “Why Should I Trust You?”: Explaining the Predictions of Any Classifier
- A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks
- A Survey on Multi-Task Learning
- Deep Learning: A Critical Appraisal
- Learning to Multitask
Cited by
Related papers
- Why I'm not Answering: An Abstention-Based Approach to Classify Cancer Pathology Reports
- The TCGA Meta-Dataset Clinical Benchmark
- Deep learning for precision medicine
- Instance Weighting for Patient-Specific Risk Stratification Models
- Machine Learning Explainability in Breast Cancer Survival
- An Optimal Learning Model for Training Expert System to Detect Uterine Cancer
- Informed Attentive Predictors: A Generalisable Architecture for Prior Knowledge-Based Assisted Diagnosis of Cancers
- Decoupling Inherent Risk and Early Cancer Signs in Image-based Breast Cancer Risk Models
- KnowRisk: An Interpretable Knowledge-Guided Model for Disease Risk Prediction