Average-case complexity in statistical inference: A puzzle-driven research seminar

2025-06-27

AI Summary

Low student engagement, superficial theoretical understanding, and suboptimal discussion quality persist in advanced graduate-level statistical inference courses. Method: This study innovatively adapts the jigsaw learning method to a graduate seminar on “average-case complexity,” restructuring traditional seminar pedagogy via modular knowledge decomposition, cross-group expert collaboration, student-led presentations with peer assessment, and structured problem-set feedback. Contribution/Results: Empirical evaluation demonstrates significant increases in both classroom participation rates and frequency of deep, conceptually grounded discussions; improved accuracy in understanding abstract constructs such as the statistical–computational gap; and the development of a validated, pedagogically tested problem set. This work constitutes the first systematic empirical validation of jigsaw learning in theory-intensive graduate seminars, offering a reusable, evidence-based instructional framework for active learning design in advanced statistics education.

Technology Category

• Machine Learning: Statistical Relational/Logic Learning
• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Practical large-scale studies of user experience

Abstract

These notes describe our experience with running a student seminar on average-case complexity in statistical inference using the jigsaw learning format at ETH Zurich in Fall of 2024. The jigsaw learning technique is an active learning technique where students work in groups on independent parts of the task and then reassemble the groups to combine all the parts together. We implemented this technique for the proofs of various recent research developments, combined with a presentation by one of the students in the beginning of the session. We describe our experience and thoughts on such a format applied in a student research seminar: including, but not limited to, higher engagement, more accessible talks by the students, and increased student participation in discussions. In the Appendix, we include all the exercises sheets for the topic, which may be of independent interest for courses on statistical-to-computational gaps and average-case complexity.

Problem

• Exploring average-case complexity in statistical inference
• Implementing jigsaw learning for student research seminars
• Assessing engagement and participation in active learning

Innovation

• Used jigsaw learning for student collaboration
• Combined student presentations with group tasks
• Included exercises on statistical-computational gaps