Research Papers

How much are we to trust a decision made by an AI algorithm? Trusting an algorithm without cause may lead to abuse, and mistrusting it may similarly lead to disuse. Trust in an AI is only desirable if it is warranted; thus, calibrating trust is critical to ensuring appropriate use. In the name of calibrating trust appropriately, AI developers should provide contracts specifying use cases in which an algorithm can and cannot be trusted. Automated explanation of AI outputs is often touted as a method by which trust can be built in the algorithm. However, automated explanations arise from algorithms themselves, so trust in these explanations is similarly only desirable if it is warranted. Developers of algorithms explaining AI outputs (xAI algorithms) should provide similar contracts, which should specify use cases in which an explanation can and cannot be trusted.

In this study, we presented a four-level cross-domain strategy for fake news detection on pre-trained models. Cross-domain text classification is a task of a model adopting a target domain by using the knowledge of the source domain. Explainability is crucial in understanding the behaviour of these complex models. A fine-tune BERT model is used to. perform cross-domain classification with several experiments using datasets from different domains. Explanatory models like Anchor, ELI5, LIME and SHAP are used to design a novel explainable approach to cross-domain levels. The experimental analysis has given an ideal pair of XAI models on different levels of cross-domain.

In pre-clinical pathology, there is a paradox between the abundance of raw data (whole slide images from many organs of many individual animals) and the lack of pixel-level slide annotations done by pathologists. Due to time constraints and requirements from regulatory authorities, diagnoses are instead stored as slide labels. Weakly supervised training is designed to take advantage of those data, and the trained models can be used by pathologists to rank slides by their probability of containing a given lesion of interest. In this work, we propose a novel contextualized eXplainable AI (XAI) framework and its application to deep learning models trained on Whole Slide Images (WSIs) in Digital Pathology. Specifically, we apply our methods to a multi-instance-learning (MIL) model, which is trained solely on slide-level labels, without the need for pixel-level annotations. We validate quantitatively our methods by quantifying the agreements of our explanations' heatmaps with pathologists' annotations, as well as with predictions from a segmentation model trained on such annotations. We demonstrate the stability of the explanations with respect to input shifts, and the fidelity with respect to increased model performance. We quantitatively evaluate the correlation between available pixel-wise annotations and explainability heatmaps. We show that the explanations on important tiles of the whole slide correlate with tissue changes between healthy regions and lesions, but do not exactly behave like a human annotator. This result is coherent with the model training strategy.

Explaining Deep Learning Hidden Neuron Activations using Concept Induction

Abhilekha Dalal, Md Kamruzzaman Sarker, Adrita Barua +1

One of the current key challenges in Explainable AI is in correctly interpreting activations of hidden neurons. It seems evident that accurate interpretations thereof would provide insights into the question what a deep learning system has internally \emph{detected} as relevant on the input, thus lifting some of the black box character of deep learning systems. The state of the art on this front indicates that hidden node activations appear to be interpretable in a way that makes sense to humans, at least in some cases. Yet, systematic automated methods that would be able to first hypothesize an interpretation of hidden neuron activations, and then verify it, are mostly missing. In this paper, we provide such a method and demonstrate that it provides meaningful interpretations. It is based on using large-scale background knowledge -- a class hierarchy of approx. 2 million classes curated from the Wikipedia Concept Hierarchy -- together with a symbolic reasoning approach called \emph{concept induction} based on description logics that was originally developed for applications in the Semantic Web field. Our results show that we can automatically attach meaningful labels from the background knowledge to individual neurons in the dense layer of a Convolutional Neural Network through a hypothesis and verification process.

Although machine learning (ML) models of AI achieve high performances in medicine, they are not free of errors. Empowering clinicians to identify incorrect model recommendations is crucial for engendering trust in medical AI. Explainable AI (XAI) aims to address this requirement by clarifying AI reasoning to support the end users. Several studies on biomedical imaging achieved promising results recently. Nevertheless, solutions for models using tabular data are not sufficient to meet the requirements of clinicians yet. This paper proposes a methodology to support clinicians in identifying failures of ML models trained with tabular data. We built our methodology on three main pillars: decomposing the feature set by leveraging clinical context latent space, assessing the clinical association of global explanations, and Latent Space Similarity (LSS) based local explanations. We demonstrated our methodology on ML-based recognition of preterm infant morbidities caused by infection. The risk of mortality, lifelong disability, and antibiotic resistance due to model failures was an open research question in this domain. We achieved to identify misclassification cases of two models with our approach. By contextualizing local explanations, our solution provides clinicians with actionable insights to support their autonomy for informed final decisions.

Interpretable ML for Imbalanced Data

Damien A. Dablain, Colin Bellinger, Bartosz Krawczyk +2

Deep learning models are being increasingly applied to imbalanced data in high stakes fields such as medicine, autonomous driving, and intelligence analysis. Imbalanced data compounds the black-box nature of deep networks because the relationships between classes may be highly skewed and unclear. This can reduce trust by model users and hamper the progress of developers of imbalanced learning algorithms. Existing methods that investigate imbalanced data complexity are geared toward binary classification, shallow learning models and low dimensional data. In addition, current eXplainable Artificial Intelligence (XAI) techniques mainly focus on converting opaque deep learning models into simpler models (e.g., decision trees) or mapping predictions for specific instances to inputs, instead of examining global data properties and complexities. Therefore, there is a need for a framework that is tailored to modern deep networks, that incorporates large, high dimensional, multi-class datasets, and uncovers data complexities commonly found in imbalanced data (e.g., class overlap, sub-concepts, and outlier instances). We propose a set of techniques that can be used by both deep learning model users to identify, visualize and understand class prototypes, sub-concepts and outlier instances; and by imbalanced learning algorithm developers to detect features and class exemplars that are key to model performance. Our framework also identifies instances that reside on the border of class decision boundaries, which can carry highly discriminative information. Unlike many existing XAI techniques which map model decisions to gray-scale pixel locations, we use saliency through back-propagation to identify and aggregate image color bands across entire classes. Our framework is publicly available at \url{https://github.com/dd1github/XAI_for_Imbalanced_Learning}

Achieving Transparency in Distributed Machine Learning with Explainable Data Collaboration

Anna Bogdanova, Akira Imakura, Tetsuya Sakurai +3

Transparency of Machine Learning models used for decision support in various industries becomes essential for ensuring their ethical use. To that end, feature attribution methods such as SHAP (SHapley Additive exPlanations) are widely used to explain the predictions of black-box machine learning models to customers and developers. However, a parallel trend has been to train machine learning models in collaboration with other data holders without accessing their data. Such models, trained over horizontally or vertically partitioned data, present a challenge for explainable AI because the explaining party may have a biased view of background data or a partial view of the feature space. As a result, explanations obtained from different participants of distributed machine learning might not be consistent with one another, undermining trust in the product. This paper presents an Explainable Data Collaboration Framework based on a model-agnostic additive feature attribution algorithm (KernelSHAP) and Data Collaboration method of privacy-preserving distributed machine learning. In particular, we present three algorithms for different scenarios of explainability in Data Collaboration and verify their consistency with experiments on open-access datasets. Our results demonstrated a significant (by at least a factor of 1.75) decrease in feature attribution discrepancies among the users of distributed machine learning.

Explainable artificial intelligence (XAI) provides explanations for not interpretable machine learning (ML) models. While many technical approaches exist, there is a lack of validation of these techniques on real-world datasets. In this work, we present a use-case of XAI: an ML model which is trained to estimate electrification rates based on mobile phone data in Senegal. The data originate from the Data for Development challenge by Orange in 2014/15. We apply two model-agnostic, local explanation techniques and find that while the model can be verified, it is biased with respect to the population density. We conclude our paper by pointing to the two main challenges we encountered during our work: data processing and model design that might be restricted by currently available XAI methods, and the importance of domain knowledge to interpret explanations.

Evaluation Metrics for Symbolic Knowledge Extracted from Machine Learning Black Boxes: A Discussion Paper

Federico Sabbatini, Roberta Calegari

As opaque decision systems are being increasingly adopted in almost any application field, issues about their lack of transparency and human readability are a concrete concern for end-users. Amongst existing proposals to associate human-interpretable knowledge with accurate predictions provided by opaque models, there are rule extraction techniques, capable of extracting symbolic knowledge out of an opaque model. However, how to assess the level of readability of the extracted knowledge quantitatively is still an open issue. Finding such a metric would be the key, for instance, to enable automatic comparison between a set of different knowledge representations, paving the way for the development of parameter autotuning algorithms for knowledge extractors. In this paper we discuss the need for such a metric as well as the criticalities of readability assessment and evaluation, taking into account the most common knowledge representations while highlighting the most puzzling issues.

Approximately 1.25 million people in the United States are treated each year for burn injuries. Precise burn injury classification is an important aspect of the medical AI field. In this work, we propose an explainable human-in-the-loop framework for improving burn ultrasound classification models. Our framework leverages an explanation system based on the LIME classification explainer to corroborate and integrate a burn expert's knowledge -- suggesting new features and ensuring the validity of the model. Using this framework, we discover that B-mode ultrasound classifiers can be enhanced by supplying textural features. More specifically, we confirm that texture features based on the Gray Level Co-occurance Matrix (GLCM) of ultrasound frames can increase the accuracy of transfer learned burn depth classifiers. We test our hypothesis on real data from porcine subjects. We show improvements in the accuracy of burn depth classification -- from ~88% to ~94% -- once modified according to our framework.

Explainable Artificial Intelligence (XAI) methods are intended to help human users better understand the decision making of an AI agent. However, many modern XAI approaches are unintuitive to end users, particularly those without prior AI or ML knowledge. In this paper, we present a novel XAI approach we call Responsibility that identifies the most responsible training example for a particular decision. This example can then be shown as an explanation: "this is what I (the AI) learned that led me to do that". We present experimental results across a number of domains along with the results of an Amazon Mechanical Turk user study, comparing responsibility and existing XAI methods on an image classification task. Our results demonstrate that responsibility can help improve accuracy for both human end users and secondary ML models.

A Framework for Understanding and Visualizing Strategies of RL Agents

Pedro Sequeira, Daniel Elenius, Jesse Hostetler +1

Recent years have seen significant advances in explainable AI as the need to understand deep learning models has gained importance with the increased emphasis on trust and ethics in AI. Comprehensible models for sequential decision tasks are a particular challenge as they require understanding not only individual predictions but a series of predictions that interact with environmental dynamics. We present a framework for learning comprehensible models of sequential decision tasks in which agent strategies are characterized using temporal logic formulas. Given a set of agent traces, we first cluster the traces using a novel embedding method that captures frequent action patterns. We then search for logical formulas that explain the agent strategies in the different clusters. We evaluate our framework on combat scenarios in StarCraft II (SC2), using traces from a handcrafted expert policy and a trained reinforcement learning agent. We implemented a feature extractor for SC2 environments that extracts traces as sequences of high-level features describing both the state of the environment and the agent's local behavior from agent replays. We further designed a visualization tool depicting the movement of units in the environment that helps understand how different task conditions lead to distinct agent behavior patterns in each trace cluster. Experimental results show that our framework is capable of separating agent traces into distinct groups of behaviors for which our approach to strategy inference produces consistent, meaningful, and easily understood strategy descriptions.