AI RESEARCH
Understanding Sample Efficiency in Predictive Coding
arXiv CS.LG
•
ArXi:2605.11911v1 Announce Type: new Predictive Coding (PC) is an influential account of cortical learning. Much of recent work has focused on comparing PC to Backpropagation (BP) to find whether PC offers any advantages. Small scale experiments show that PC enables learning that is sample efficient and effective in many contexts, though a thorough theoretical understanding of the phenomena remains elusive.