Explainable AI: A Review of Machine Learning Interpretability Methods

Recent advances in artificial intelligence (AI) have led to its widespread industrial adoption, with machine learning systems demonstrating superhuman performance in a significant number of tasks. However, this surge in performance, has often been achieved through increased model complexity, turning such systems into “black box” approaches and causing uncertainty regarding the way they operate and, ultimately, the way that they come to decisions. This ambiguity has made it problematic for machine learning systems to be adopted in sensitive yet critical domains, where their value could be immense, such as healthcare. As a result, scientific interest in the field of Explainable Artificial Intelligence (XAI), a field that is concerned with the development of new methods that explain and interpret machine learning models, has been tremendously reignited over recent years. This study focuses on machine learning interpretability methods; more specifically, a literature review and taxonomy of these methods are presented, as well as links to their programming implementations, in the hope that this survey would serve as a reference point for both theorists and practitioners.

Download Full-text

XHAC

10.4018/978-1-7998-4186-9.ch008 ◽

2022 ◽

pp. 146-164

Author(s):

Duygu Bagci Das ◽

Derya Birant

Keyword(s):

Artificial Intelligence ◽

Machine Learning ◽

Internet Of Things ◽

Window Size ◽

Classification Performance ◽

Black Box ◽

Data Exploration ◽

Learning Models ◽

Explainable Artificial Intelligence ◽

Machine Learning Models

Explainable artificial intelligence (XAI) is a concept that has emerged and become popular in recent years. Even interpretation in machine learning models has been drawing attention. Human activity classification (HAC) systems still lack interpretable approaches. In this study, an approach, called eXplainable HAC (XHAC), was proposed in which the data exploration, model structure explanation, and prediction explanation of the ML classifiers for HAR were examined to improve the explainability of the HAR models' components such as sensor types and their locations. For this purpose, various internet of things (IoT) sensors were considered individually, including accelerometer, gyroscope, and magnetometer. The location of these sensors (i.e., ankle, arm, and chest) was also taken into account. The important features were explored. In addition, the effect of the window size on the classification performance was investigated. According to the obtained results, the proposed approach makes the HAC processes more explainable compared to the black-box ML techniques.

Download Full-text

Learning to Validate the Predictions of Black Box Machine Learning Models on Unseen Data

Proceedings of the Workshop on Human-In-the-Loop Data Analytics - HILDA'19 ◽

10.1145/3328519.3329126 ◽

2019 ◽

Author(s):

Sergey Redyuk ◽

Sebastian Schelter ◽

Tammo Rukat ◽

Volker Markl ◽

Felix Biessmann

Keyword(s):

Machine Learning ◽

Black Box ◽

Learning Models ◽

Unseen Data ◽

Machine Learning Models

Download Full-text

Uncovering and Correcting Shortcut Learning in Machine Learning Models for Skin Cancer Diagnosis

Diagnostics ◽

10.3390/diagnostics12010040 ◽

2021 ◽

Vol 12 (1) ◽

pp. 40

Author(s):

Meike Nauta ◽

Ricky Walsh ◽

Adam Dubowski ◽

Christin Seifert

Keyword(s):

Machine Learning ◽

Clinical Practice ◽

Skin Cancer ◽

Cancer Diagnosis ◽

Image Inpainting ◽

Relevant Information ◽

Black Box ◽

Training Dataset ◽

Learning Models ◽

Machine Learning Models

Machine learning models have been successfully applied for analysis of skin images. However, due to the black box nature of such deep learning models, it is difficult to understand their underlying reasoning. This prevents a human from validating whether the model is right for the right reasons. Spurious correlations and other biases in data can cause a model to base its predictions on such artefacts rather than on the true relevant information. These learned shortcuts can in turn cause incorrect performance estimates and can result in unexpected outcomes when the model is applied in clinical practice. This study presents a method to detect and quantify this shortcut learning in trained classifiers for skin cancer diagnosis, since it is known that dermoscopy images can contain artefacts. Specifically, we train a standard VGG16-based skin cancer classifier on the public ISIC dataset, for which colour calibration charts (elliptical, coloured patches) occur only in benign images and not in malignant ones. Our methodology artificially inserts those patches and uses inpainting to automatically remove patches from images to assess the changes in predictions. We find that our standard classifier partly bases its predictions of benign images on the presence of such a coloured patch. More importantly, by artificially inserting coloured patches into malignant images, we show that shortcut learning results in a significant increase in misdiagnoses, making the classifier unreliable when used in clinical practice. With our results, we, therefore, want to increase awareness of the risks of using black box machine learning models trained on potentially biased datasets. Finally, we present a model-agnostic method to neutralise shortcut learning by removing the bias in the training dataset by exchanging coloured patches with benign skin tissue using image inpainting and re-training the classifier on this de-biased dataset.

Download Full-text

Train Deep Learning Models using subsurface geological images datasets

10.5194/egusphere-egu21-6385 ◽

2021 ◽

Author(s):

Ramy Abdallah ◽

Clare E. Bond ◽

Robert W.H. Butler

Keyword(s):

Artificial Intelligence ◽

Machine Learning ◽

Neural Networks ◽

Deep Learning ◽

Convolutional Neural Networks ◽

Learning Models ◽

Structural Interpretation ◽

Performance Accuracy ◽

Learning Techniques ◽

Machine Learning Models

<p>Machine learning is being presented as a new solution for a wide range of geoscience problems. Primarily machine learning has been used for 3D seismic data processing, seismic facies analysis and well log data correlation. The rapid development in technology with open-source artificial intelligence libraries and the accessibility of affordable computer graphics processing units (GPU) makes the application of machine learning in geosciences increasingly tractable. However, the application of artificial intelligence in structural interpretation workflows of subsurface datasets is still ambiguous. This study aims to use machine learning techniques to classify images of folds and fold-thrust structures. Here we show that convolutional neural networks (CNNs) as supervised deep learning techniques provide excellent algorithms to discriminate between geological image datasets. Four different datasets of images have been used to train and test the machine learning models. These four datasets are a seismic character dataset with five classes (faults, folds, salt, flat layers and basement), folds types with three classes (buckle, chevron and conjugate), fault types with three classes (normal, reverse and thrust) and fold-thrust geometries with three classes (fault bend fold, fault propagation fold and detachment fold). These image datasets are used to investigate three machine learning models. One Feedforward linear neural network model and two convolutional neural networks models (Convolution 2d layer transforms sequential model and Residual block model (ResNet with 9, 34, and 50 layers)). Validation and testing datasets forms a critical part of testing the model&#8217;s performance accuracy. The ResNet model records the highest performance accuracy score, of the machine learning models tested. Our CNN image classification model analysis provides a framework for applying machine learning to increase structural interpretation efficiency, and shows that CNN classification models can be applied effectively to geoscience problems. The study provides a starting point to apply unsupervised machine learning approaches to sub-surface structural interpretation workflows.</p>

Download Full-text

Recent Progress in Quantum Machine Learning

Limitations and Future Applications of Quantum Cryptography - Advances in Information Security, Privacy, and Ethics ◽

10.4018/978-1-7998-6677-0.ch012 ◽

2021 ◽

pp. 232-256

Author(s):

Amandeep Singh Bhatia ◽

Renata Wong

Keyword(s):

Artificial Intelligence ◽

Machine Learning ◽

Quantum Computing ◽

Recent Progress ◽

Machine Learning Algorithms ◽

Learning Models ◽

Exciting Field ◽

Quantum Machine Learning ◽

The Subject ◽

Machine Learning Models

Quantum computing is a new exciting field which can be exploited to great speed and innovation in machine learning and artificial intelligence. Quantum machine learning at crossroads explores the interaction between quantum computing and machine learning, supplementing each other to create models and also to accelerate existing machine learning models predicting better and accurate classifications. The main purpose is to explore methods, concepts, theories, and algorithms that focus and utilize quantum computing features such as superposition and entanglement to enhance the abilities of machine learning computations enormously faster. It is a natural goal to study the present and future quantum technologies with machine learning that can enhance the existing classical algorithms. The objective of this chapter is to facilitate the reader to grasp the key components involved in the field to be able to understand the essentialities of the subject and thus can compare computations of quantum computing with its counterpart classical machine learning algorithms.

Download Full-text

Research Methods to Study and Empower Crowd Workers

10.1093/oso/9780198860679.003.0009 ◽

2021 ◽

pp. 164-184

Author(s):

Saiph Savage ◽

Carlos Toxtli ◽

Eber Betanzos-Torres

Keyword(s):

Artificial Intelligence ◽

Machine Learning ◽

Research Methods ◽

Real World ◽

Intelligent Systems ◽

Quantitative Information ◽

Learning Models ◽

Professional Goals ◽

Career Growth ◽

Machine Learning Models

The artificial intelligence (AI) industry has created new jobs that are essential to the real world deployment of intelligent systems. Part of the job focuses on labelling data for machine learning models or having workers complete tasks that AI alone cannot do. These workers are usually known as ‘crowd workers’—they are part of a large distributed crowd that is jointly (but separately) working on the tasks although they are often invisible to end-users, leading to workers often being paid below minimum wage and having limited career growth. In this chapter, we draw upon the field of human–computer interaction to provide research methods for studying and empowering crowd workers. We present our Computational Worker Leagues which enable workers to work towards their desired professional goals and also supply quantitative information about crowdsourcing markets. This chapter demonstrates the benefits of this approach and highlights important factors to consider when researching the experiences of crowd workers.

Download Full-text

Unsupervised by any other name: Hidden layers of knowledge production in artificial intelligence on social media

Big Data & Society ◽

10.1177/2053951718819569 ◽

2019 ◽

Vol 6 (1) ◽

pp. 205395171881956 ◽

Cited By ~ 10

Author(s):

Anja Bechmann ◽

Geoffrey C Bowker

Keyword(s):

Artificial Intelligence ◽

Machine Learning ◽

Knowledge Production ◽

Classification Theory ◽

Human Knowledge ◽

Learning Models ◽

Work Process ◽

Learning Development ◽

High Degree ◽

Machine Learning Models

Artificial Intelligence (AI) in the form of different machine learning models is applied to Big Data as a way to turn data into valuable knowledge. The rhetoric is that ensuing predictions work well—with a high degree of autonomy and automation. We argue that we need to analyze the process of applying machine learning in depth and highlight at what point human knowledge production takes place in seemingly autonomous work. This article reintroduces classification theory as an important framework for understanding such seemingly invisible knowledge production in the machine learning development and design processes. We suggest a framework for studying such classification closely tied to different steps in the work process and exemplify the framework on two experiments with machine learning applied to Facebook data from one of our labs. By doing so we demonstrate ways in which classification and potential discrimination take place in even seemingly unsupervised and autonomous models. Moving away from concepts of non-supervision and autonomy enable us to understand the underlying classificatory dispositifs in the work process and that this form of analysis constitutes a first step towards governance of artificial intelligence.

Download Full-text