Siyi (Carrie) Gu

Hospital-acquired pressure injury (HAPI), a key indicator of nursing quality, is associated with adverse outcomes and economic burdens on patients. Thus, predicting HAPI is important. While models using structured data have been proposed, unstructured notes are rarely used yet contain useful patient information. Here we evaluate the impact of using various natural language processing packages to predict HAPI using clinical notes. To discover important information from unstructured text, we use named entity recognition (NER) packages to identify keywords. These keywords are used to construct the feature space of our classifier for HAPI prediction. We compare three different NER tools on MIMIC-III and assess the impact of vocabulary size reduction by comparing the use of all clinical notes with only nursing notes. Our results suggest that NER extraction using nursing notes can yield accurate models while providing interpretable keywords that play a significant role in the prediction of HAPI.

As the societal impact of Deep Neural Networks (DNNs) grows, the goals for advancing DNNs become more complex and diverse, ranging from improving a conventional model accuracy metric to infusing advanced human virtues such as fairness, accountability, transparency (FaccT), and unbiasedness. Recently, techniques in Explainable Artificial Intelligence (XAI) are attracting considerable attention, and have tremendously helped Machine Learning (ML) engineers in understanding AI models. However, at the same time, we started to witness the emerging need beyond XAI among AI communities; based on the insights learned from XAI, how can we better empower ML engineers in steering their DNNs so that the model’s reasonableness and performance can be improved as intended? This article provides a timely and extensive literature overview of the field Explanation-Guided Learning (EGL), a domain of techniques that steer the DNNs’ reasoning process by adding regularization, supervision, or intervention on model explanations. In doing so, we first provide a formal definition of EGL and its general learning paradigm. Secondly, an overview of the key factors for EGL evaluation, as well as summarization and categorization of existing evaluation procedures and metrics for EGL are provided. Finally, the current and potential future application areas and directions of EGL are discussed, and an extensive experimental study is presented aiming at providing comprehensive comparative studies among existing EGL models in various popular application domains, such as Computer Vision (CV) and Natural Language Processing (NLP) domains.

Despite the fast progress of explanation techniques in modern Deep Neural Networks (DNNs) where the main focus is handling “how to generate the explanations”, advanced research questions that exam- ine the quality of the explanation itself (e.g., “whether the explana- tions are accurate”) and improve the explanation quality (e.g., “how to adjust the model to generate more accurate explanations when explanations are inaccurate”) are still relatively under-explored. To guide the model toward better explanations, techniques in expla- nation supervision—which add supervision signals on the model explanation—have started to show promising effects on improving both the generalizability as and intrinsic interpretability of Deep Neural Networks. However, the research on supervising explana- tions, especially in vision-based applications represented through saliency maps, is in its early stage due to several inherent challenges: 1) inaccuracy of the human explanation annotation boundary, 2) incompleteness of the human explanation annotation region, and 3) inconsistency of the data distribution between human annota- tion and model explanation maps. To address the challenges, we propose a generic RES framework for guiding visual explanation by developing a novel objective that handles inaccurate boundary, incomplete region, and inconsistent distribution of human anno- tations, with a theoretical justification on model generalizability. Extensive experiments on two real-world image datasets demon- strate the effectiveness of the proposed framework on enhancing both the reasonability of the explanation and the performance of the backbone DNNs model.