Research Papers

Filling gaps in trustworthy development of AI

Shahar Avin, Haydn Belfield, Miles Brundage +9

The range of application of artificial intelligence (AI) is vast, as is the potential for harm. Growing awareness of potential risks from AI systems has spurred action to address those risks, while eroding confidence in AI systems and the organizations that develop them. A 2019 study found over 80 organizations that published and adopted "AI ethics principles'', and more have joined since. But the principles often leave a gap between the "what" and the "how" of trustworthy AI development. Such gaps have enabled questionable or ethically dubious behavior, which casts doubts on the trustworthiness of specific organizations, and the field more broadly. There is thus an urgent need for concrete methods that both enable AI developers to prevent harm and allow them to demonstrate their trustworthiness through verifiable behavior. Below, we explore mechanisms (drawn from arXiv:2004.07213) for creating an ecosystem where AI developers can earn trust - if they are trustworthy. Better assessment of developer trustworthiness could inform user choice, employee actions, investment decisions, legal recourse, and emerging governance regimes.

Learning from class imbalanced datasets poses challenges for many machine learning algorithms. Many real-world domains are, by definition, class imbalanced by virtue of having a majority class that naturally has many more instances than its minority class (e.g. genuine bank transactions occur much more often than fraudulent ones). Many methods have been proposed to solve the class imbalance problem, among the most popular being oversampling techniques (such as SMOTE). These methods generate synthetic instances in the minority class, to balance the dataset, performing data augmentations that improve the performance of predictive machine learning (ML) models. In this paper we advance a novel data augmentation method (adapted from eXplainable AI), that generates synthetic, counterfactual instances in the minority class. Unlike other oversampling techniques, this method adaptively combines exist-ing instances from the dataset, using actual feature-values rather than interpolating values between instances. Several experiments using four different classifiers and 25 datasets are reported, which show that this Counterfactual Augmentation method (CFA) generates useful synthetic data points in the minority class. The experiments also show that CFA is competitive with many other oversampling methods many of which are variants of SMOTE. The basis for CFAs performance is discussed, along with the conditions under which it is likely to perform better or worse in future tests.

The problem of individualization is recognized as crucial in almost every field. Identifying causes of effects in specific events is likewise essential for accurate decision making. However, such estimates invoke counterfactual relationships, and are therefore indeterminable from population data. For example, the probability of benefiting from a treatment concerns an individual having a favorable outcome if treated and an unfavorable outcome if untreated. Experiments conditioning on fine-grained features are fundamentally inadequate because we can't test both possibilities for an individual. Tian and Pearl provided bounds on this and other probabilities of causation using a combination of experimental and observational data. Even though those bounds were proven tight, narrower bounds, sometimes significantly so, can be achieved when structural information is available in the form of a causal model. This has the power to solve central problems, such as explainable AI, legal responsibility, and personalized medicine, all of which demand counterfactual logic. We analyze and expand on existing research by applying bounds to the probability of necessity and sufficiency (PNS) along with graphical criteria and practical applications.

Conceptual understanding through efficient inverse-design of quantum optical experiments

Mario Krenn, Jakob Kottmann, Nora Tischler +1

One crucial question within artificial intelligence research is how this technology can be used to discover new scientific concepts and ideas. We present Theseus, an explainable AI algorithm that can contribute to science at a conceptual level. This work entails four significant contributions. (i) We introduce an interpretable representation of quantum optical experiments amenable to algorithmic use. (ii) We develop an inverse-design approach for new quantum experiments, which is orders of magnitudes faster than the best previous methods. (iii) We solve several crucial open questions in quantum optics, which is expected to advance photonic technology. Finally, and most importantly, (iv) the interpretable representation and drastic speedup produce solutions that a human scientist can interpret outright to discover new scientific concepts. We anticipate that Theseus will become an essential tool in quantum optics and photonic hardware, with potential applicability to other quantum physical disciplines.

Work on "learning with rationales" shows that humans providing explanations to a machine learning system can improve the system's predictive accuracy. However, this work has not been connected to work in "explainable AI" which concerns machines explaining their reasoning to humans. In this work, we show that learning with rationales can also improve the quality of the machine's explanations as evaluated by human judges. Specifically, we present experiments showing that, for CNN- based text classification, explanations generated using "supervised attention" are judged superior to explanations generated using normal unsupervised attention.

Private Accountability in the Age of Artificial Intelligence

Sonia Katyal

In this Article, I explore the impending conflict between the protection of civil rights and artificial intelligence (AI). While both areas of law have amassed rich and well-developed areas of scholarly work and doctrinal support, a growing body of scholars are interrogating the intersection between them. This Article argues that the issues surrounding algorithmic accountability demonstrate a deeper, more structural tension within a new generation of disputes regarding law and technology. As I argue, the true promise of AI does not lie in the information we reveal to one another, but rather in the questions it raises about the interaction of technology, property, and civil rights. For this reason, I argue that we are looking in the wrong place if we look only to the state to address issues of algorithmic accountability. Instead, we must turn to other ways to ensure more transparency and accountability that stem from private industry, rather than public regulation. The issue of algorithmic bias represents a crucial new world of civil rights concerns, one that is distinct in nature from the ones that preceded it. Since we are in a world where the activities of private corporations, rather than the state, are raising concerns about privacy, due process, and discrimination, we must focus on the role of private corporations in addressing the issue. Towards this end, I discuss a variety of tools to help eliminate the opacity of AI, including codes of conduct, impact statements, and whistleblower protection, which I argue carries the potential to encourage greater endogeneity in civil rights enforcement. Ultimately, by examining the relationship between private industry and civil rights, we can perhaps develop a new generation of forms of accountability in the process.

Phishing emails continue to pose a significant threat, causing financial losses and security breaches. This study addresses limitations in existing research, such as reliance on proprietary datasets and lack of real-world application, by proposing a high-performance machine learning model for email classification. Utilizing a comprehensive and largest available public dataset, the model achieves a f1 score of 0.99 and is designed for deployment within relevant applications. Additionally, Explainable AI (XAI) is integrated to enhance user trust. This research offers a practical and highly accurate solution, contributing to the fight against phishing by empowering users with a real-time web-based application for phishing email detection.

This research paper focuses on Acute Lymphoblastic Leukemia (ALL), a form of blood cancer prevalent in children and teenagers, characterized by the rapid proliferation of immature white blood cells (WBCs). These atypical cells can overwhelm healthy cells, leading to severe health consequences. Early and accurate detection of ALL is vital for effective treatment and improving survival rates. Traditional diagnostic methods are time-consuming, costly, and prone to errors. The paper proposes an automated detection approach using computer-aided diagnostic (CAD) models, leveraging deep learning techniques to enhance the accuracy and efficiency of leukemia diagnosis. The study utilizes various transfer learning models like ResNet101V2, VGG19, InceptionV3, and InceptionResNetV2 for classifying ALL. The methodology includes using the Local Interpretable Model-Agnostic Explanations (LIME) for ensuring the validity and reliability of the AI system's predictions. This approach is critical for overcoming the "black box" nature of AI, where decisions made by models are often opaque and unaccountable. The paper highlights that the proposed method using the InceptionV3 model achieved an impressive 98.38% accuracy, outperforming other tested models. The results, verified by the LIME algorithm, showcase the potential of this method in accurately identifying ALL, providing a valuable tool for medical practitioners. The research underscores the impact of explainable artificial intelligence (XAI) in medical diagnostics, paving the way for more transparent and trustworthy AI applications in healthcare.

Benefiting from the rapid development of deep learning, 2D and 3D computer vision applications are deployed in many safe-critical systems, such as autopilot and identity authentication. However, deep learning models are not trustworthy enough because of their limited robustness against adversarial attacks. The physically realizable adversarial attacks further pose fatal threats to the application and human safety. Lots of papers have emerged to investigate the robustness and safety of deep learning models against adversarial attacks. To lead to trustworthy AI, we first construct a general threat model from different perspectives and then comprehensively review the latest progress of both 2D and 3D adversarial attacks. We extend the concept of adversarial examples beyond imperceptive perturbations and collate over 170 papers to give an overview of deep learning model robustness against various adversarial attacks. To the best of our knowledge, we are the first to systematically investigate adversarial attacks for 3D models, a flourishing field applied to many real-world applications. In addition, we examine physical adversarial attacks that lead to safety violations. Last but not least, we summarize present popular topics, give insights on challenges, and shed light on future research on trustworthy AI.

CX-ToM: Counterfactual Explanations with Theory-of-Mind for Enhancing Human Trust in Image Recognition Models

Arjun R. Akula, Keze Wang, Changsong Liu +5

We propose CX-ToM, short for counterfactual explanations with theory-of mind, a new explainable AI (XAI) framework for explaining decisions made by a deep convolutional neural network (CNN). In contrast to the current methods in XAI that generate explanations as a single shot response, we pose explanation as an iterative communication process, i.e. dialog, between the machine and human user. More concretely, our CX-ToM framework generates sequence of explanations in a dialog by mediating the differences between the minds of machine and human user. To do this, we use Theory of Mind (ToM) which helps us in explicitly modeling human's intention, machine's mind as inferred by the human as well as human's mind as inferred by the machine. Moreover, most state-of-the-art XAI frameworks provide attention (or heat map) based explanations. In our work, we show that these attention based explanations are not sufficient for increasing human trust in the underlying CNN model. In CX-ToM, we instead use counterfactual explanations called fault-lines which we define as follows: given an input image I for which a CNN classification model M predicts class c_pred, a fault-line identifies the minimal semantic-level features (e.g., stripes on zebra, pointed ears of dog), referred to as explainable concepts, that need to be added to or deleted from I in order to alter the classification category of I by M to another specified class c_alt. We argue that, due to the iterative, conceptual and counterfactual nature of CX-ToM explanations, our framework is practical and more natural for both expert and non-expert users to understand the internal workings of complex deep learning models. Extensive quantitative and qualitative experiments verify our hypotheses, demonstrating that our CX-ToM significantly outperforms the state-of-the-art explainable AI models.

The use of deep neural networks (DNNs) in safety-critical applications like mobile health and autonomous driving is challenging due to numerous model-inherent shortcomings. These shortcomings are diverse and range from a lack of generalization over insufficient interpretability to problems with malicious inputs. Cyber-physical systems employing DNNs are therefore likely to suffer from safety concerns. In recent years, a zoo of state-of-the-art techniques aiming to address these safety concerns has emerged. This work provides a structured and broad overview of them. We first identify categories of insufficiencies to then describe research activities aiming at their detection, quantification, or mitigation. Our paper addresses both machine learning experts and safety engineers: The former ones might profit from the broad range of machine learning topics covered and discussions on limitations of recent methods. The latter ones might gain insights into the specifics of modern ML methods. We moreover hope that our contribution fuels discussions on desiderata for ML systems and strategies on how to propel existing approaches accordingly.

A Study of Automatic Metrics for the Evaluation of Natural Language Explanations

Miruna Clinciu, Arash Eshghi, Helen Hastie

As transparency becomes key for robotics and AI, it will be necessary to evaluate the methods through which transparency is provided, including automatically generated natural language (NL) explanations. Here, we explore parallels between the generation of such explanations and the much-studied field of evaluation of Natural Language Generation (NLG). Specifically, we investigate which of the NLG evaluation measures map well to explanations. We present the ExBAN corpus: a crowd-sourced corpus of NL explanations for Bayesian Networks. We run correlations comparing human subjective ratings with NLG automatic measures. We find that embedding-based automatic NLG evaluation methods, such as BERTScore and BLEURT, have a higher correlation with human ratings, compared to word-overlap metrics, such as BLEU and ROUGE. This work has implications for Explainable AI and transparent robotic and autonomous systems.