Related papers: Dynamic Models of Learning and Education Measureme…

Meta-learning involves training models on a variety of training tasks in a way that enables them to generalize well on new, unseen test tasks. In this work, we consider meta-learning within the framework of high-dimensional multivariate…

Normalization methods are a central building block in the deep learning toolbox. They accelerate and stabilize training, while decreasing the dependence on manually tuned learning rate schedules. When learning from multi-modal…

Standard reinforcement learning (RL) optimizes policies for reward but imposes few constraints on how decisions evolve over time. As a result, policies may achieve high performance while exhibiting temporally incoherent behavior such as…

Several years ago I reported a survey (Hake 1998a,b,c) of pre/post test data for 62 introductory physics courses enrolling a total of 6542 students. The present article provides a summary of that survey and presents fourteen lessons from…

Virtually all machine learning tasks are characterized using some form of loss function, and "good performance" is typically stated in terms of a sufficiently small average loss, taken over the random draw of test data. While optimizing for…

Model selection is treated as a standard performance boosting step in many machine learning applications. Once all other properties of a learning problem are fixed, the model is selected by grid search on a held-out validation set. This is…

Regression models usually tend to recover a noisy signal in the form of a combination of regressors, also called features in machine learning, themselves being the result of a learning process.The alignment of the prior covariance feature…

Through in-class observations of teaching assistants (TAs) and students in the lab sections of a large introductory physics course, we study which TA behaviors can be used to predict student engagement and, in turn, how this engagement…

Measuring learning progress is essential for curiosity-driven exploration in reinforcement learning, but widely used signals such as prediction error often fail to distinguish meaningful, learnable patterns from random noise. This paper…

This study examines the role of AI-assisted pretesting in enhancing learning outcomes, particularly when integrated with generative AI tools like ChatGPT. Pretesting, a learning strategy in which students attempt to answer questions or…

We introduce and theoretically justify a new measure of the self-recognition component of student physics identity called Self-Evaluated Expertise (SEE). This measure is constructed such that it can be extracted from existing responses to…

Conceptual tests are widely used by physics instructors to assess students' conceptual understanding and compare teaching methods. It is common to look at students' changes in their answers between a pre-test and a post-test to quantify a…

Considering learner engagement has a mutual benefit for both learners and instructors. Instructors can help learners increase their attention, involvement, motivation, and interest. On the other hand, instructors can improve their…

Instruction tuning (IT) achieves impressive zero-shot generalization results by training large language models (LLMs) on a massive amount of diverse tasks with instructions. However, how to select new tasks to improve the performance and…

Grokking is the phenomenon whereby, unlike the training performance, which peaks early in the training process, the test/generalization performance of a model stagnates over arbitrarily many epochs and then suddenly jumps to usually close…

The mainstream paradigm behind continual learning has been to adapt the model parameters to non-stationary data distributions, where catastrophic forgetting is the central challenge. Typical methods rely on a rehearsal buffer or known task…

We propose a physics-informed EDFA gain model based on the active learning method. Experimental results show that the proposed modelling method can reach a higher optimal accuracy and reduce ~90% training data to achieve the same…

Typically, loss functions, regularization mechanisms and other important aspects of training parametric models are chosen heuristically from a limited set of options. In this paper, we take the first step towards automating this process,…

The research presented in this thesis was motivated by the need to improve introductory physics courses. Introductory physics courses are generally the first courses in which students learn to create models to solve complex problems.…

Continual learning is motivated by the need to adapt to real-world dynamics in tasks and data distribution while mitigating catastrophic forgetting. Despite significant advances in continual learning techniques, the theoretical…