Why Code, Why Now: Learnability, Computability, and the Real Limits of Machine Learning
This work addresses a foundational problem in machine learning by challenging the assumption that scaling alone will solve remaining challenges, which is significant for researchers and practitioners in AI.
The paper tackles the problem of understanding why code generation scales more reliably than reinforcement learning by proposing a five-level hierarchy of learnability based on information structure, arguing that ML progress is limited by task learnability rather than model size.
Code generation has progressed more reliably than reinforcement learning, largely because code has an information structure that makes it learnable. Code provides dense, local, verifiable feedback at every token, whereas most reinforcement learning problems do not. This difference in feedback quality is not binary but graded. We propose a five-level hierarchy of learnability based on information structure and argue that the ceiling on ML progress depends less on model size than on whether a task is learnable at all. The hierarchy rests on a formal distinction among three properties of computational problems (expressibility, computability, and learnability). We establish their pairwise relationships, including where implications hold and where they fail, and present a unified template that makes the structural differences explicit. The analysis suggests why supervised learning on code scales predictably while reinforcement learning does not, and why the common assumption that scaling alone will solve remaining ML challenges warrants scrutiny.