AlphaGrad: Non-Linear Gradient Normalization Optimizer

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: Sane, Soham
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866913804056526848
author Sane, Soham
author_facet Sane, Soham
contents We introduce AlphaGrad, a memory-efficient, conditionally stateless optimizer addressing the memory overhead and hyperparameter complexity of adaptive methods like Adam. AlphaGrad enforces scale invariance via tensor-wise L2 gradient normalization followed by a smooth hyperbolic tangent transformation, $g' = \tanh(α\cdot \tilde{g})$, controlled by a single steepness parameter $α$. Our contributions include: (1) the AlphaGrad algorithm formulation; (2) a formal non-convex convergence analysis guaranteeing stationarity; (3) extensive empirical evaluation on diverse RL benchmarks (DQN, TD3, PPO). Compared to Adam, AlphaGrad demonstrates a highly context-dependent performance profile. While exhibiting instability in off-policy DQN, it provides enhanced training stability with competitive results in TD3 (requiring careful $α$ tuning) and achieves substantially superior performance in on-policy PPO. These results underscore the critical importance of empirical $α$ selection, revealing strong interactions between the optimizer's dynamics and the underlying RL algorithm. AlphaGrad presents a compelling alternative optimizer for memory-constrained scenarios and shows significant promise for on-policy learning regimes where its stability and efficiency advantages can be particularly impactful.
format Preprint
id arxiv_https___arxiv_org_abs_2504_16020
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AlphaGrad: Non-Linear Gradient Normalization Optimizer
Sane, Soham
Machine Learning
Artificial Intelligence
Neural and Evolutionary Computing
We introduce AlphaGrad, a memory-efficient, conditionally stateless optimizer addressing the memory overhead and hyperparameter complexity of adaptive methods like Adam. AlphaGrad enforces scale invariance via tensor-wise L2 gradient normalization followed by a smooth hyperbolic tangent transformation, $g' = \tanh(α\cdot \tilde{g})$, controlled by a single steepness parameter $α$. Our contributions include: (1) the AlphaGrad algorithm formulation; (2) a formal non-convex convergence analysis guaranteeing stationarity; (3) extensive empirical evaluation on diverse RL benchmarks (DQN, TD3, PPO). Compared to Adam, AlphaGrad demonstrates a highly context-dependent performance profile. While exhibiting instability in off-policy DQN, it provides enhanced training stability with competitive results in TD3 (requiring careful $α$ tuning) and achieves substantially superior performance in on-policy PPO. These results underscore the critical importance of empirical $α$ selection, revealing strong interactions between the optimizer's dynamics and the underlying RL algorithm. AlphaGrad presents a compelling alternative optimizer for memory-constrained scenarios and shows significant promise for on-policy learning regimes where its stability and efficiency advantages can be particularly impactful.
title AlphaGrad: Non-Linear Gradient Normalization Optimizer
topic Machine Learning
Artificial Intelligence
Neural and Evolutionary Computing
url https://arxiv.org/abs/2504.16020