Deceptive agents are a challenge for the safety, trustworthiness, and cooperation of AI systems.

Fox News contributor Jonathan Morris and Pastor Robert Jeffress react to the president unveiling new guidance on public school prayer. The battle over prayer in school is raging in Texas right now, with Attorney General Ken Paxton vowing to defend any school district that introduces the controversial practice under a recent state law expanding religious expression in education. For the entirety of my life, and I'm old, the prohibition on public school-sponsored prayer seemed like settled Constitutional science, owing to a 1962 Supreme Court decision barring what had previously been a widespread and normal practice. In the past, I agreed with this form of separation of church and state. For me it was almost a question of better safe than sorry regarding the rights of minority religions, and importantly, I believed that Christian moral values were so ingrained in our culture that 30 seconds a day of praying could be forsaken.

In a recent paper, Kit Fine presents some striking results concerning the logical properties of (first-order) ignorance, second-order ignorance and Rumsfeld ignorance. However, Rumsfeld ignorance is definable in terms of ignorance, which makes some existing results and the axiomatization problem trivial. A main reason is that the accessibility relations for the implicit knowledge operator contained in the packaged operators of ignorance and Rumsfeld ignorance are the same. In this work, we assume the two accessibility relations to be different so that one of them is an arbitrary subset of the other. This will avoid the definability issue and retain most of the previous validities. The main results are axiomatizations over various proper bi-frame classes. Finally we apply our framework to analyze Fine's results.

Informational parsimony provides a useful inductive bias for learning representations that achieve better generalization by being robust to noise and spurious correlations. We propose information gating as a way to learn parsimonious representations that identify the minimal information required for a task. When gating information, we can learn to reveal as little information as possible so that a task remains solvable, or hide as little information as possible so that a task becomes unsolvable. We gate information using a differentiable parameterization of the signal-to-noise ratio, which can be applied to arbitrary values in a network, e.g., erasing pixels at the input layer or activations in some intermediate layer. When gating at the input layer, our models learn which visual cues matter for a given task.

The Neyman-Pearson region of a simple binary hypothesis testing is the set of points whose coordinates represent the false positive rate and false negative rate of some test. The lower boundary of this region is given by the Neyman-Pearson lemma, and is up to a coordinate change, equivalent to the optimal ROC curve. We establish a novel lower bound for the boundary in terms of any $f$-divergence. Since the bound generated by hockey-stick $f$-divergences characterizes the Neyman-Pearson boundary, this bound is best possible. In the case of KL divergence, this bound improves Pinsker's inequality. Furthermore, we obtain a closed-form refined upper bound for the Neyman-Pearson boundary in terms of the Chernoff $α$-coefficient. Finally, we present methods for constructing pairs of distributions that can approximately or exactly realize any given Neyman-Pearson boundary.

While analysing challenges in pilot projects developing AI with marginalized communities, we found it difficult to express them within commonly used paradigms. We therefore constructed an alternative conceptual framework to ground AI development in the social fabric -- the Cloud Weaving Model -- inspired (amongst others) by indigenous knowledge, motifs from nature, and Eastern traditions. This paper introduces and elaborates on the fundamental elements of the model (clouds, spiders, threads, spiderwebs, and weather) and their interpretation in an AI context. The framework is then applied to comprehend patterns observed in co-creation pilots approaching marginalized communities, highlighting neglected yet relevant dimensions for responsible AI development.

Informational parsimony provides a useful inductive bias for learning representations that achieve better generalization by being robust to noise and spurious correlations. We propose information gating as a way to learn parsimonious representations that identify the minimal information required for a task. When gating information, we can learn to reveal as little information as possible so that a task remains solvable, or hide as little information as possible so that a task becomes unsolvable. We gate information using a differentiable parameterization of the signal-to-noise ratio, which can be applied to arbitrary values in a network, e.g., erasing pixels at the input layer or activations in some intermediate layer. When gating at the input layer, our models learn which visual cues matter for a given task.

This paper proposes a new measure of fairness for classification and regression problems based on welfare considerations rather than inequality considerations. This measure of fairness represents a convex constraint, making it easy to optimize for. They experimentally demonstrate the tradeoffs between this notion of fairness and previous notions. I believe this to be a pretty valuable submission. A welfare-based approach over a inequality-based approach should turn out to be very helpful in addressing all sorts of concerns with the current literature. It also provokes a number of questions to follow up on which, while disappointing that they are not addressed here, means that the community should take interest in this paper.

The authors restrict their analysis to aleatoric and epistemic uncertainty (leaving out numerical uncertainty). Aleatoric uncertainty includes the uncertainty from statistical noise in data. Epistemic uncertainty is usually another term for ignorance, i.e. things one could in theory know but doesn't know in practice. In this paper however, the authors use epistemic uncertainty as a synonym for model uncertainty (or structural uncertainty, i.e. ignorance over the true physical model which created the data). The paper raises the question how these two sources of uncertainty can be jointly quantified, and when which source of uncertainty is dominant.

Natural revision seems so natural: it changes beliefs as little as possible to incorporate new information. Yet, some counterexamples show it wrong. It is so conservative that it never fully believes. It only believes in the current conditions. This is right in some cases and wrong in others. Which is which? The answer requires extending natural revision from simple formulae expressing universal truths (something holds) to conditionals expressing conditional truth (something holds in certain conditions). The extension is based on the basic principles natural revision follows, identified as minimal change, indifference and naivety: change beliefs as little as possible; equate the likeliness of scenarios by default; believe all until contradicted. The extension says that natural revision restricts changes to the current conditions. A comparison with an unrestricting revision shows what exactly the current conditions are. It is not what currently considered true if it contradicts the new information. It includes something more and more unlikely until the new information is at least possible.