paper-with-me

Papers

Narrowing the Focus: Learned Optimizers for Pretrained Models

2024-08-17 · Gus Kristiansen, Mark Sandler, Andrey Zhmoginov, Nolan Miller, Anirudh Goyal, JIhwan Lee, Max Vladymyrov

In modern deep learning, the models are learned by applying gradient updates using an optimizer, which transforms the updates based on various statistics. Optimizers are often hand-designed and tuning their hyperparameters is a big part of the training process. Learned optimizers have shown some initial promise, but are generally unsuccessful as a general optimization mechanism applicable to every problem. In this work we explore a different direction: instead of learning general optimizers, we instead specialize them to a specific training environment. We propose a novel optimizer technique that learns a layer-specific linear combination of update directions provided by a set of base optimizers, effectively adapting its strategy to the specific model and dataset. When evaluated on image classification tasks, this specialized optimizer significantly outperforms both traditional off-the-shelf methods such as Adam, as well as existing general learned optimizers. Moreover, it demonstrates robust generalization with respect to model initialization, evaluating on unseen datasets, and training durations beyond its meta-training horizon.

📄 PDF Abstract BibTeX arXiv:2408.09310

Code (0)

등록된 구현이 없습니다.

Tasks

image-classificationImage Classification

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically
BASE 설명 없음
Adam 설명 없음

Similar Papers 제목 키워드 기반

Overcoming barriers to the training of effective learned optimizers

2021-01-01 · Luke Metz, Niru Maheswaranathan, C. Daniel Freeman, Ben Poole 외

Much as replacing hand-designed features with learned functions has revolutionized how we solve perceptual tasks, we believe learned algorithms will transform how we train models. In this work we focus on general-purpose…

Tasks, stability, architecture, and compute: Training more effective learned optimizers, and using them to train themselves

2020-09-23 · Luke Metz, Niru Maheswaranathan, C. Daniel Freeman, Ben Poole 외

Much as replacing hand-designed features with learned functions has revolutionized how we solve perceptual tasks, we believe learned algorithms will transform how we train models. In this work we focus on general-purpose…

PyLO: Towards Accessible Learned Optimizers in PyTorch

2025-06-12 · Paul Janson, Benjamin Therien, Quentin Anthony, Xiaolong Huang 외

Learned optimizers have been an active research topic over the past decade, with increasing progress toward practical, general-purpose optimizers that can serve as drop-in replacements for widely used methods like Adam. …

Reverse engineering learned optimizers reveals known and novel mechanisms

2020-11-04 · NeurIPS 2021 12 · Niru Maheswaranathan, David Sussillo, Luke Metz, Ruoxi Sun 외

Learned optimizers are algorithms that can themselves be trained to solve optimization problems. In contrast to baseline optimizers (such as momentum or Adam) that use simple update rules derived from theoretical princip…

Data-Driven Performance Guarantees for Classical and Learned Optimizers

2024-04-22 · Rajiv Sambharya, Bartolomeo Stellato

We introduce a data-driven approach to analyze the performance of continuous optimization algorithms using generalization guarantees from statistical learning theory. We study classical and learned optimizers to solve fa…

Learning TheoryMeta-Learning