# Break a Lag: Triple Exponential Moving Average for Enhanced Optimization

Source: [Break a Lag: Triple Exponential Moving Average for Enhanced Optimization](<https://www.sota2.com/research/paper/68fa762d1e209131cea6bb05>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-27T06:44:57.967Z

Authors: Roi Peleg, Yair Smadar, Teddy Lazebnik, Assaf Hoogi

Published: 2023-06-02T10:29:33.000Z

## Abstract

The performance of deep learning models is critically dependent on sophisticated optimization strategies. While existing optimizers have shown promising results, many rely on first-order Exponential Moving Average (EMA) techniques, which often limit their ability to track complex gradient trends accurately. This fact can lead to a significant lag in trend identification and suboptimal optimization, particularly in highly dynamic gradient behavior. To address this fundamental limitation, we introduce Fast Adaptive Moment Estimation (FAME), a novel optimization technique that leverages the power of Triple Exponential Moving Average. By incorporating an advanced tracking mechanism, FAME enhances responsiveness to data dynamics, mitigates trend identification lag, and optimizes learning efficiency. Our comprehensive evaluation encompasses different computer vision tasks including image classification, object detection, and semantic segmentation, integrating FAME into 30 distinct architectures ranging from lightweight CNNs to Vision Transformers. Through rigorous benchmarking against state-of-the-art optimizers, FAME demonstrates superior accuracy and robustness. Notably, it offers high scalability, delivering substantial improvements across diverse model complexities, architectures, tasks, and benchmarks.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2306.01423>)

[PDF](<https://arxiv.org/pdf/2306.01423.pdf>)