Gary S. Collins

Lena Maier-Hein, Annika Reinke, Patrick Godau et al. · utoronto

Increasing evidence shows that flaws in machine learning (ML) algorithm validation are an underestimated global problem. Particularly in automatic biomedical image analysis, chosen performance metrics often do not reflect the domain interest, thus failing to adequately measure scientific progress and hindering translation of ML techniques into practice. To overcome this, our large international expert consortium created Metrics Reloaded, a comprehensive framework guiding researchers in the problem-aware selection of metrics. Following the convergence of ML methodology across application domains, Metrics Reloaded fosters the convergence of validation methodology. The framework was developed in a multi-stage Delphi process and is based on the novel concept of a problem fingerprint - a structured representation of the given problem that captures all aspects that are relevant for metric selection, from the domain interest to the properties of the target structure(s), data set and algorithm output. Based on the problem fingerprint, users are guided through the process of choosing and applying appropriate validation metrics while being made aware of potential pitfalls. Metrics Reloaded targets image analysis problems that can be interpreted as a classification task at image, object or pixel level, namely image-level classification, object detection, semantic segmentation, and instance segmentation tasks. To improve the user experience, we implemented the framework in the Metrics Reloaded online tool, which also provides a point of access to explore weaknesses, strengths and specific recommendations for the most common validation metrics. The broad applicability of our framework across domains is demonstrated by an instantiation for various biological and medical image analysis use cases.

Annika Reinke, Minu D. Tizabi, Michael Baumgartner et al.

Validation metrics are key for the reliable tracking of scientific progress and for bridging the current chasm between artificial intelligence (AI) research and its translation into practice. However, increasing evidence shows that particularly in image analysis, metrics are often chosen inadequately in relation to the underlying research problem. This could be attributed to a lack of accessibility of metric-related knowledge: While taking into account the individual strengths, weaknesses, and limitations of validation metrics is a critical prerequisite to making educated choices, the relevant knowledge is currently scattered and poorly accessible to individual researchers. Based on a multi-stage Delphi process conducted by a multidisciplinary expert consortium as well as extensive community feedback, the present work provides the first reliable and comprehensive common point of access to information on pitfalls related to validation metrics in image analysis. Focusing on biomedical image analysis but with the potential of transfer to other fields, the addressed pitfalls generalize across application domains and are categorized according to a newly created, domain-agnostic taxonomy. To facilitate comprehension, illustrations and specific examples accompany each pitfall. As a structured body of information accessible to researchers of all levels of expertise, this work enhances global comprehension of a key topic in image analysis validation.

Giovanni E. Cacciamani, Michael B. Eppler, Conner Ganjavi et al.

The swift progress and ubiquitous adoption of Generative AI (GAI), Generative Pre-trained Transformers (GPTs), and large language models (LLMs) like ChatGPT, have spurred queries about their ethical application, use, and disclosure in scholarly research and scientific productions. A few publishers and journals have recently created their own sets of rules; however, the absence of a unified approach may lead to a 'Babel Tower Effect,' potentially resulting in confusion rather than desired standardization. In response to this, we present the ChatGPT, Generative Artificial Intelligence, and Natural Large Language Models for Accountable Reporting and Use Guidelines (CANGARU) initiative, with the aim of fostering a cross-disciplinary global inclusive consensus on the ethical use, disclosure, and proper reporting of GAI/GPT/LLM technologies in academia. The present protocol consists of four distinct parts: a) an ongoing systematic review of GAI/GPT/LLM applications to understand the linked ideas, findings, and reporting standards in scholarly research, and to formulate guidelines for its use and disclosure, b) a bibliometric analysis of existing author guidelines in journals that mention GAI/GPT/LLM, with the goal of evaluating existing guidelines, analyzing the disparity in their recommendations, and identifying common rules that can be brought into the Delphi consensus process, c) a Delphi survey to establish agreement on the items for the guidelines, ensuring principled GAI/GPT/LLM use, disclosure, and reporting in academia, and d) the subsequent development and dissemination of the finalized guidelines and their supplementary explanation and elaboration documents.

Jared Markowitz, Jesse Silverberg, Gary Collins · mila

By reusing data throughout training, off-policy deep reinforcement learning algorithms offer improved sample efficiency relative to on-policy approaches. For continuous action spaces, the most popular methods for off-policy learning include policy improvement steps where a learned state-action ($Q$) value function is maximized over selected batches of data. These updates are often paired with regularization to combat associated overestimation of $Q$ values. With an eye toward safety, we revisit this strategy in environments with "mixed-sign" reward functions; that is, with reward functions that include independent positive (incentive) and negative (cost) terms. This setting is common in real-world applications, and may be addressed with or without constraints on the cost terms. We find the combination of function approximation and a term that maximizes $Q$ in the policy update to be problematic in such environments, because systematic errors in value estimation impact the contributions from the competing terms asymmetrically. This results in overemphasis of either incentives or costs and may severely limit learning. We explore two remedies to this issue. First, consistent with prior work, we find that periodic resetting of $Q$ and policy networks can be used to reduce value estimation error and improve learning in this setting. Second, we formulate novel off-policy actor-critic methods for both unconstrained and constrained learning that do not explicitly maximize $Q$ in the policy update. We find that this second approach, when applied to continuous action spaces with mixed-sign rewards, consistently and significantly outperforms state-of-the-art methods augmented by resetting. We further find that our approach produces agents that are both competitive with popular methods overall and more reliably competent on frequently-studied control problems that do not have mixed-sign rewards.

Ben Van Calster, Gary S. Collins, Andrew J. Vickers et al.

A myriad of measures to illustrate performance of predictive artificial intelligence (AI) models have been proposed in the literature. Selecting appropriate performance measures is essential for predictive AI models that are developed to be used in medical practice, because poorly performing models may harm patients and lead to increased costs. We aim to assess the merits of classic and contemporary performance measures when validating predictive AI models for use in medical practice. We focus on models with a binary outcome. We discuss 32 performance measures covering five performance domains (discrimination, calibration, overall, classification, and clinical utility) along with accompanying graphical assessments. The first four domains cover statistical performance, the fifth domain covers decision-analytic performance. We explain why two key characteristics are important when selecting which performance measures to assess: (1) whether the measure's expected value is optimized when it is calculated using the correct probabilities (i.e., a "proper" measure), and (2) whether they reflect either purely statistical performance or decision-analytic performance by properly considering misclassification costs. Seventeen measures exhibit both characteristics, fourteen measures exhibited one characteristic, and one measure possessed neither characteristic (the F1 measure). All classification measures (such as classification accuracy and F1) are improper for clinically relevant decision thresholds other than 0.5 or the prevalence. We recommend the following measures and plots as essential to report: AUROC, calibration plot, a clinical utility measure such as net benefit with decision curve analysis, and a plot with probability distributions per outcome category.

Sebastian Vollmer, Bilal A. Mateen, Gergo Bohner et al.

Machine learning (ML), artificial intelligence (AI) and other modern statistical methods are providing new opportunities to operationalize previously untapped and rapidly growing sources of data for patient benefit. Whilst there is a lot of promising research currently being undertaken, the literature as a whole lacks: transparency; clear reporting to facilitate replicability; exploration for potential ethical concerns; and, clear demonstrations of effectiveness. There are many reasons for why these issues exist, but one of the most important that we provide a preliminary solution for here is the current lack of ML/AI- specific best practice guidance. Although there is no consensus on what best practice looks in this field, we believe that interdisciplinary groups pursuing research and impact projects in the ML/AI for health domain would benefit from answering a series of questions based on the important issues that exist when undertaking work of this nature. Here we present 20 questions that span the entire project life cycle, from inception, data analysis, and model evaluation, to implementation, as a means to facilitate project planning and post-hoc (structured) independent evaluation. By beginning to answer these questions in different settings, we can start to understand what constitutes a good answer, and we expect that the resulting discussion will be central to developing an international consensus framework for transparent, replicable, ethical and effective research in artificial intelligence (AI-TREE) for health.