Skip to content
conftrace_

maximal update parameterization

4 papers

Explore in graph

Also known as

ΜP MUP

Co-occurring keywords · papers mentioning both

neural network optimization (2) feature learning (2) neural network scaling (1) infinite-width limit (1) target propagation (1) spectral scaling (1) scaling rule (1) early training stage (1) local loss optimization (1) macro-variables (1)

Papers

Local Loss Optimization in the Infinite Width: Stable Parameterization of Predictive Coding Networks and Target Propagation ICLR 2025 On the Provable Separation of Scales in Maximal Update Parameterization ICML 2025 On Feature Learning in Structured State Space Models NeurIPS 2024 On the Parameterization of Second-Order Optimization Effective towards the Infinite Width ICLR 2024