AI RESEARCH
Free Random Projection for In-Context Reinforcement Learning
arXiv CS.LG
•
ArXi:2504.06983v3 Announce Type: replace Hierarchical inductive biases are hypothesized to promote generalizable policies in reinforcement learning, as nstrated by explicit hyperbolic latent representations and architectures. Therefore, a flexible approach is to have these biases emerge naturally from the algorithm. We