The 38th International Conference on Algorithmic Learning Theory (ALT 2027)

Its scope covers the design and analysis of learning algorithms; statistical, computational and information-theoretic foundations; online and multi-agent learning and game theory; convex and nonconvex optimisation, implicit bias and overparameterisation; reinforcement learning including LLM post-training; theory of transformers and large language models; privacy, fairness, alignment and robustness; distribution shift; approximation and generalisation in deep learning; high-dimensional, nonparametric and causal statistics; and the analysis of sampling methods such as diffusion models. Accepted papers are presented as full-length talks and published in the Proceedings of Machine Learning Research. The conference is held in person and does not support remote talks.