While highly accurate, the resulting predictors are large, slow, and opaque as compared to their constituents.

Language is an interface to the outside world.

First provide a summary of the paper, and then address the following criteria: Quality, clarity, originality and significance. This paper proposes a probabilistic approach for learning the assignment of exercises to skills from student data, where student knowledge changes while exercises are being solved; the model also estimates the student knowledge while estimating the skill assignments. The paper uses a weighted CRP to model the assignment, incorporating expert labelings through the weighting. In simulation, the method recovers skill labelings with high accuracy, with little dependence on the expert labels, and across several datasets, the paper finds that skill labelings from this method result in higher prediction accuracy than other approaches. Overall, I found the paper to be clear and the proposed model is a relatively novel extension of existing methods.

Neural Information Processing Systems http://nips.cc/

A good teacher can adjust a curriculum based on students' learning history.

Neural Information Processing Systems http://nips.cc/

Then, the goal of the agent is to maximize cumulative rewards over time by identifying an optimal action which has the maximum reward. However, since MABs often assume that prior knowledge about rewards is not given, the agent faces an innate dilemma between gathering new information by exploring sub-optimal actions (exploration) and choosing the best action based on the collected information (exploitation). Designing an efficient exploration algorithm for MABs is a long-standing challenging problem.

Diffusion models have achieved state-of-the-art performance, demonstrating remarkable generalisation capabilities across diverse domains. However, the mechanisms underpinning these strong capabilities remain only partially understood. A leading conjecture, based on the manifold hypothesis, attributes this success to their ability to adapt to low-dimensional geometric structure within the data. This work provides evidence for this conjecture, focusing on how such phenomena could result from the formulation of the learning problem through score matching. We inspect the role of implicit regularisation by investigating the effect of smoothing minimisers of the empirical score matching objective. Our theoretical and empirical results confirm that smoothing the score function -- or equivalently, smoothing in the log-density domain -- produces smoothing tangential to the data manifold. In addition, we show that the manifold along which the diffusion model generalises can be controlled by choosing an appropriate smoothing.