Explainability of Deep Neural Networks
Type
Master’s thesis / Bachelor’s thesis / guided research
Prerequisites
- Knowledge of deep learning with image data, natural language data, or graph data
- Proficiency with Python and deep learning frameworks (either PyTorch or Tensorflow)
Description
Over the last decade, deep learning methods have been deployed in numerous real-world, often safety-critical, applications. However, a major and growing concern remains the explainability of neural network decisions. A neural network operates as a black box: Apriori, one can only comprehend the input and output of a neural net decision, not the reasoning leading to the decision. The explainable AI (XAI) field aims to develop explanation methods that "open the black box" and shed light on the reasoning behind neural network decisions.
References
- A Rate-Distortion Framework for Explaining Black-box Model Decisions (https://arxiv.org/abs/2110.08252)
- Cartoon Explanations of Image Classifiers (https://arxiv.org/abs/2110.03485)
- In-distribution Interpretability for Challenging Modalities (https://arxiv.org/pdf/2007.00758.pdf)
- Axiomatic Attribution for Deep Networks (https://arxiv.org/abs/1703.01365)
- On Pixel-Wise Explanations for Non-Linear Classifier Decisions by Layer-Wise Relevance Propagation (https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0130140)
- A Unified Approach to Interpreting Model Predictions (https://arxiv.org/pdf/1705.07874.pdf)
- The (un)reliability of saliency methods (https://arxiv.org/pdf/1711.00867.pdf)
- Explaining Deep Neural Networks and Beyond: A Review of Methods and Applications (https://arxiv.org/pdf/2003.07631.pdf)