[2004.05107] Levels of Analysis for Machine Learning

Machine learning is currently involved in some of the most vigorous debates it has ever seen. Such debates often seem to go around in circles, reaching no conclusion or resolution. This is perhaps unsurprising given that researchers in machine learning come to these discussions with very different frames of reference, making it challenging for them to align perspectives and find common ground. As a remedy for this dilemma, we advocate for the adoption of a common conceptual framework which can be used to understand, analyze, and discuss research. We present one such framework which is popular in cognitive science and neuroscience and which we believe has great utility in machine learning as well: Marr’s levels of analysis. Through a series of case studies, we demonstrate how the levels facilitate an understanding and dissection of several methods from machine learning. By adopting the levels of analysis in one’s own work, we argue that researchers can be better equipped to engage in th

Levels of Analysis for Machine Learning Jessica B. Hamrick DeepMind jhamrick@google.com &Shakir Mohamed DeepMind shakir@google.com Abstract Machine learning is currently involved in some of the most vigorous debates it has ever seen. Such debates often seem to go around in circles, reaching no conclusion or resolution. This is perhaps unsurprising given that researchers in machine learning come to these discussions with very different frames of reference, making it challenging for them to align perspectives and find common ground. As a remedy for this dilemma, we advocate for the adoption of a

saved by

• Andria Xu

related reading

• NL.pdfabehrouz.github.io
• [2604.21691] There Will Be a Scientific Theory of Deep Learningarxiv.org
• The Building Blocks of Interpretabilitydistill.pub
• [2604.21691] There Will Be a Scientific Theory of Deep Learningarxiv.org
• [2510.13786] The Art of Scaling Reinforcement Learning Compute for LLMsarxiv.org
• Understanding deep learning requires rethinking generalizationarxiv.org
• [2403.02327] Model Lakesarxiv.org
• [1709.06560] Deep Reinforcement Learning that Mattersarxiv.org
• [1707.01415] Machine Learning, Deepest Learning: Statistical Data Assimilation Problemsarxiv.org
• A short note on interpretability and mindsericjmichaud.com
• Fact Finding: Attempting to Reverse-Engineer Factual Recall on the Neuron Level (Post 1) — AI Alignment Forumalignmentforum.org
• On the Learnability of Physical Concepts: Can a Neural Network Understand What’s Real?arxiv.org