paper-with-me

Papers

MADA: Meta-Adaptive Optimizers through hyper-gradient Descent

2024-01-17 · Kaan Ozkara, Can Karakus, Parameswaran Raman, Mingyi Hong, Shoham Sabach, Branislav Kveton, Volkan Cevher

Following the introduction of Adam, several novel adaptive optimizers for deep learning have been proposed. These optimizers typically excel in some tasks but may not outperform Adam uniformly across all tasks. In this work, we introduce Meta-Adaptive Optimizers (MADA), a unified optimizer framework that can generalize several known optimizers and dynamically learn the most suitable one during training. The key idea in MADA is to parameterize the space of optimizers and dynamically search through it using hyper-gradient descent during training. We empirically compare MADA to other popular optimizers on vision and language tasks, and find that MADA consistently outperforms Adam and other popular optimizers, and is robust against sub-optimally tuned hyper-parameters. MADA achieves a greater validation performance improvement over Adam compared to other popular optimizers during GPT-2 training and fine-tuning. We also propose AVGrad, a modification of AMSGrad that replaces the maximum operator with averaging, which is more suitable for hyper-gradient optimization. Finally, we provide a convergence analysis to show that parameterized interpolations of optimizers can improve their error bounds (up to constants), hinting at an advantage for meta-optimizers.

📄 PDF Abstract BibTeX arXiv:2401.08893

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Cosine Annealing Cosine Annealing is a type of learning rate schedule that has the effect of starting with a large learning rate that is relatively rapidly decreased to a minimum value before…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Residual Connection 설명 없음
Average Pooling 설명 없음
Linear Warmup With Cosine Annealing Linear Warmup With Cosine Annealing is a learning rate schedule where we increase the learning rate linearly for $n$ updates and then anneal according to a cosine schedule…
Convolution A convolution is a type of matrix operation, consisting of a kernel, a small matrix of weights, that slides over input data performing element-wise multiplication with the…
Weight Decay 설명 없음

Similar Papers 제목 키워드 기반

When Can You Get Away with Low Memory Adam?

2025-03-03 · Dayal Singh Kalra, John Kirchenbauer, Maissam Barkeshli, Tom Goldstein

Adam is the go-to optimizer for training modern machine learning models, but it requires additional memory to maintain the moving averages of the gradients and their squares. While various low-memory optimizers have been…

Combining Optimization Methods Using an Adaptive Meta Optimizer

2021-06-19 · Algorithsm MDPI 2021 6 · Nicola Landro, Ignazio Gallo, Riccardo La Grassa

Optimization methods are of great importance for the efficient training of neural networks. There are many articles in the literature that propose particular variants of existing optimizers. In our article, we propose th…

Articles

Learning Versatile Optimizers on a Compute Diet

2025-01-22 · Abhinav Moudgil, Boris Knyazev, Guillaume Lajoie, Eugene Belilovsky

Learned optimization has emerged as a promising alternative to hand-crafted optimizers, with the potential to discover stronger learned update rules that enable faster, hyperparameter-free training of neural networks. A …

GPU

MOCA-HESP: Meta High-dimensional Bayesian Optimization for Combinatorial and Mixed Spaces via Hyper-ellipsoid Partitioning

2025-08-09 · Lam Ngo, Huong Ha, Jeffrey Chan, Hongyu Zhang arxiv

High-dimensional Bayesian Optimization (BO) has attracted significant attention in recent research. However, existing methods have mainly focused on optimizing in continuous domains, while combinatorial (ordinal and cate…

Bidirectional Looking with A Novel Double Exponential Moving Average to Adaptive and Non-adaptive Momentum Optimizers

2023-07-02 · Yineng Chen, Zuchao Li, Lefei Zhang, Bo Du 외

Optimizer is an essential component for the success of deep learning, which guides the neural network to update the parameters according to the loss on the training set. SGD and Adam are two classical and effective optim…