Research Papers

Incorporating Expert Guidance in Epidemic Forecasting

Alexander Rodríguez, Bijaya Adhikari, Naren Ramakrishnan +1

Forecasting influenza like illnesses (ILI) has rapidly progressed in recent years from an art to a science with a plethora of data-driven methods. While these methods have achieved qualified success, their applicability is limited due to their inability to incorporate expert feedback and guidance systematically into the forecasting framework. We propose a new approach leveraging the Seldonian optimization framework from AI safety and demonstrate how it can be adapted to epidemic forecasting. We study two types of guidance: smoothness and regional consistency of errors, where we show that by its successful incorporation, we are able to not only bound the probability of undesirable behavior to happen, but also to reduce RMSE on test data by up to 17%.

We examine the problem of explainable AI (xAI) and explore what delivering xAI means in practice, particularly in contexts that involve formal or informal and ad-hoc collaboration where agency and accountability in decision-making are achieved and sustained interactionally. We use an example from an earlier study of collaborative decision-making in screening mammography and the difficulties users faced when trying to interpret the behavior of an AI tool to illustrate the challenges of delivering usable and effective xAI. We conclude by setting out a study programme for future research to help advance our understanding of xAI requirements for safe and ethical AI.

Is a new regulated profession, such as Artificial Intelligence (AI) Architect who is responsible and accountable for AI outputs necessary to ensure trustworthy AI? AI is becoming all pervasive and is often deployed in everyday technologies, devices and services without our knowledge. There is heightened awareness of AI in recent years which has brought with it fear. This fear is compounded by the inability to point to a trustworthy source of AI, however even the term "trustworthy AI" itself is troublesome. Some consider trustworthy AI to be that which complies with relevant laws, while others point to the requirement to comply with ethics and standards (whether in addition to or in isolation of the law). This immediately raises questions of whose ethics and which standards should be applied and whether these are sufficient to produce trustworthy AI in any event.

In February 2020, the Montreal AI Ethics Institute (MAIEI) was invited by the Office of the Privacy Commissioner of Canada (OPCC) to provide for comments both at a closed roundtable and in writing on the OPCC consultation proposal for amendments relative to Artificial Intelligence (AI), to the Canadian privacy legislation, the Personal Information Protection and Electronic Documents Act (PIPEDA). The present document includes MAIEI comments and recommendations in writing. Per MAIEI's mission and mandate to act as a catalyst for public feedback pertaining to AI Ethics and regulatory technology developments, as well as to provide for public competence-building workshops on critical topics in such domains, the reader will also find such public feedback and propositions by Montrealers who participated at MAIEI's workshops, submitted as Schedule 1 to the present report. For each of OPCC 12 proposals, and underlying questions, as described on its website, MAIEI provides a short reply, a summary list of recommendations, as well as comments relevant to the question at hand. We leave you with three general statements to keep in mind while going through the next pages: 1) AI systems should be used to augment human capacity for meaningful and purposeful connections and associations, not as a substitute for trust. 2) Humans have collectively accepted to uphold the rule of law, but for machines, the code is rule. Where socio-technical systems are deployed to make important decisions, profiles or inferences about individuals, we will increasingly have to attempt the difficult exercise of drafting and encoding our law in a manner learnable by machines. 3) Let us work collectively towards a world where Responsible AI becomes the rule, before our socio-technical systems become "too connected to fail".

Montreal AI Ethics Institute's Response to Scotland's AI Strategy

Abhishek Gupta

In January and February 2020, the Scottish Government released two documents for review by the public regarding their artificial intelligence (AI) strategy. The Montreal AI Ethics Institute (MAIEI) reviewed these documents and published a response on 4 June 2020. MAIEI's response examines several questions that touch on the proposed definition of AI; the people-centered nature of the strategy; considerations to ensure that everyone benefits from AI; the strategy's overarching vision; Scotland's AI ecosystem; the proposed strategic themes; and how to grow public confidence in AI by building responsible and ethical systems. In addition to examining the points above, MAIEI suggests that the strategy be extended to include considerations on biometric data and how that will be processed and used in the context of AI. It also highlights the importance of tackling head-on the inherently stochastic nature of deep learning systems and developing concrete guidelines to ensure that these systems are built responsibly and ethically, particularly as machine learning becomes more accessible. Finally, it concludes that any national AI strategy must clearly address the measurements of success in regards to the strategy's stated goals and vision to ensure that they are interpreted and applied consistently. To do this, there must be inclusion and transparency between those building the systems and those using them in their work.

The overarching goal of Explainable AI is to develop systems that not only exhibit intelligent behaviours, but also are able to explain their rationale and reveal insights. In explainable machine learning, methods that produce a high level of prediction accuracy as well as transparent explanations are valuable. In this work, we present an explainable classification method. Our method works by first constructing a symbolic Knowledge Base from the training data, and then performing probabilistic inferences on such Knowledge Base with linear programming. Our approach achieves a level of learning performance comparable to that of traditional classifiers such as random forests, support vector machines and neural networks. It identifies decisive features that are responsible for a classification as explanations and produces results similar to the ones found by SHAP, a state of the art Shapley Value based method. Our algorithms perform well on a range of synthetic and non-synthetic data sets.

Machine learning software is increasingly being used to make decisions that affect people's lives. But sometimes, the core part of this software (the learned model), behaves in a biased manner that gives undue advantages to a specific group of people (where those groups are determined by sex, race, etc.). This "algorithmic discrimination" in the AI software systems has become a matter of serious concern in the machine learning and software engineering community. There have been works done to find "algorithmic bias" or "ethical bias" in the software system. Once the bias is detected in the AI software system, the mitigation of bias is extremely important. In this work, we a)explain how ground-truth bias in training data affects machine learning model fairness and how to find that bias in AI software,b)propose a methodFairwaywhich combines pre-processing and in-processing approach to remove ethical bias from training data and trained model. Our results show that we can find bias and mitigate bias in a learned model, without much damaging the predictive performance of that model. We propose that (1) test-ing for bias and (2) bias mitigation should be a routine part of the machine learning software development life cycle. Fairway offers much support for these two purposes.

TX-Ray: Quantifying and Explaining Model-Knowledge Transfer in (Un-)Supervised NLP

Nils Rethmeier, Vageesh Kumar Saxena, Isabelle Augenstein

While state-of-the-art NLP explainability (XAI) methods focus on explaining per-sample decisions in supervised end or probing tasks, this is insufficient to explain and quantify model knowledge transfer during (un-)supervised training. Thus, for TX-Ray, we modify the established computer vision explainability principle of 'visualizing preferred inputs of neurons' to make it usable transfer analysis and NLP. This allows one to analyze, track and quantify how self- or supervised NLP models first build knowledge abstractions in pretraining (1), and then transfer these abstractions to a new domain (2), or adapt them during supervised fine-tuning (3). TX-Ray expresses neurons as feature preference distributions to quantify fine-grained knowledge transfer or adaptation and guide human analysis. We find that, similar to Lottery Ticket based pruning, TX-Ray based pruning can improve test set generalization and that it can reveal how early stages of self-supervision automatically learn linguistic abstractions like parts-of-speech.

The social implications of algorithmic decision-making in sensitive contexts have generated lively debates among multiple stakeholders, such as moral and political philosophers, computer scientists, and the public. Yet, the lack of a common language and a conceptual framework for an appropriate bridging of the moral, technical, and political aspects of the debate prevents the discussion to be as effective as it can be. Social scientists and psychologists are contributing to this debate by gathering a wealth of empirical data, yet a philosophical analysis of the social implications of algorithmic decision-making remains comparatively impoverished. In attempting to address this lacuna, this paper argues that a hierarchy of different types of explanations for why and how an algorithmic decision outcome is achieved can establish the relevant connection between the moral and technical aspects of algorithmic decision-making. In particular, I offer a multi-faceted conceptual framework for the explanations and the interpretations of algorithmic decisions, and I claim that this framework can lay the groundwork for a focused discussion among multiple stakeholders about the social implications of algorithmic decision-making, as well as AI governance and ethics more generally.

Artificial intelligence (AI) is becoming increasingly widespread in system development endeavors. As AI systems affect various stakeholders due to their unique nature, the growing influence of these systems calls for ethical considerations. Academic discussion and practical examples of autonomous system failures have highlighted the need for implementing ethics in software development. However, research on methods and tools for implementing ethics into AI system design and development in practice is still lacking. This paper begins to address this focal problem by providing elements needed for producing a baseline for ethics in AI based software development. We do so by means of an industrial multiple case study on AI systems development in the healthcare sector. Using a research model based on extant, conceptual AI ethics literature, we explore the current state of practice out on the field in the absence of formal methods and tools for ethically aligned design.

General purpose language models are increasingly able to control robotic hardware. Understanding the capabilities and safety of these models when embodied is therefore increasingly important for understanding their societal impact and risks. To this end, we introduce Inspect Robots, a modular, open-source framework for developing and running evaluations of embodied agents. Inspect Robots pairs customizable, reusable abstractions for specifying physical evaluations and analyzing their results with infrastructure that automates evaluation setup, execution and termination. We demonstrate Inspect Robots by using it to evaluate the capabilities and safety of six policies based on frontier language models. Inspect Robots has seen significant early uptake, receiving nearly 100,000 downloads in the three months since its release.

AI reviewers can assign different judgments to manuscripts that report the same science in different wording, potentially rewarding rhetorical optimization over scientific improvement. We formulate Rhetorical Robustness as the joint requirement of stability across content-preserving rewrites and discrimination across papers. We introduce RobustReview, a controlled full-manuscript benchmark with 1,260 manuscript versions, and evaluate 30 reviewer configurations. The benchmark reveals false robustness, where low rewrite sensitivity coincides with score collapse across papers, and shows that human alignment and rhetorical robustness rank reviewers differently. Moreover, the evaluated content-focused prompting protocol does not consistently improve robustness across backbones. Motivated by these findings, we introduce SciCore, a dual-branch reviewer that averages a full-manuscript judgment with a judgment based on an extracted, structured science core. This design combines manuscript-level assessment with a content-normalized view intended to reduce rhetorical sensitivity. In our primary GPT-5.5 comparison, SciCore achieves a leading joint stability-discrimination profile among the benchmarked reviewers while maintaining competitive human alignment. These results identify rhetorical robustness as a distinct evaluation target and demonstrate the potential of science-core review to improve it.