Subject
3 entries
Meta Learning
Bookmarks
In-Context Reinforcement Learning with Algorithm Distillation
Algorithm Distillation trains a causal transformer on sequences of RL learning histories so the model can improve its policy entirely in-context without gradient updates. A key step toward meta-learning agents that get better at RL through experience rather than parameter updates.
Self-Programming Artificial Intelligence
This paper explores the concept of self-programming AI — systems that can inspect, modify, or write their own code and learning algorithms. It sits at the intersection of meta-learning and program synthesis, asking whether AI systems can improve their own architecture or training procedure through learned introspection rather than human-designed updates.
An Open Source AutoML Benchmark
This paper presents an open-source, extensible benchmark for comparing AutoML systems across 39 classification datasets, finding that no single system consistently dominates a tuned random forest baseline. It establishes best practices for fair AutoML evaluation and provides a living framework that accepts community contributions.