SMMF: Square-Matricized Momentum Factorization for Memory-Efficient Optimization
- URL: http://arxiv.org/abs/2412.08894v2
- Date: Fri, 13 Dec 2024 04:03:14 GMT
- Title: SMMF: Square-Matricized Momentum Factorization for Memory-Efficient Optimization
- Authors: Kwangryeol Park, Seulki Lee,
- Abstract summary: SMMF is a memory-efficient that reduces the memory requirement of the widely used adaptive learning rate Matrix, such as Adam, by up to 96%.
We conduct a regret bound analysis of SMMF, which shows that it converges similarly to non-memory-efficient adaptive learning rate Matrix, such as AdamNC.
In our experiment, SMMF takes up to 96% less memory compared to state-of-the-art memory efficients, e.g., Adafactor, CAME, and SM3, while achieving comparable model performance.
- Score: 0.5755004576310332
- License:
- Abstract: We propose SMMF (Square-Matricized Momentum Factorization), a memory-efficient optimizer that reduces the memory requirement of the widely used adaptive learning rate optimizers, such as Adam, by up to 96%. SMMF enables flexible and efficient factorization of an arbitrary rank (shape) of the first and second momentum tensors during optimization, based on the proposed square-matricization and one-time single matrix factorization. From this, it becomes effectively applicable to any rank (shape) of momentum tensors, i.e., bias, matrix, and any rank-d tensors, prevalent in various deep model architectures, such as CNNs (high rank) and Transformers (low rank), in contrast to existing memory-efficient optimizers that applies only to a particular (rank-2) momentum tensor, e.g., linear layers. We conduct a regret bound analysis of SMMF, which shows that it converges similarly to non-memory-efficient adaptive learning rate optimizers, such as AdamNC, providing a theoretical basis for its competitive optimization capability. In our experiment, SMMF takes up to 96% less memory compared to state-of-the-art memory efficient optimizers, e.g., Adafactor, CAME, and SM3, while achieving comparable model performance on various CNN and Transformer tasks.
Related papers
- Regularized second-order optimization of tensor-network Born machines [2.8834278113855896]
Born machines (TNBMs) are quantum-inspired generative models for learning data distributions.
We present an improved second-order optimization technique for TNBM training, which significantly enhances convergence rates and the quality of the optimized model.
arXiv Detail & Related papers (2025-01-30T19:00:04Z) - Tensor-GaLore: Memory-Efficient Training via Gradient Tensor Decomposition [93.98343072306619]
We present Navier-GaLore, a novel method for efficient training of neural networks with higher-order tensor weights.
Across various PDE tasks, Navier-GaLore achieves substantial memory savings, reducing memory usage by up to 75%.
arXiv Detail & Related papers (2025-01-04T20:51:51Z) - Expanding Sparse Tuning for Low Memory Usage [103.43560327427647]
We propose a method named SNELL (Sparse tuning with kerNELized LoRA) for sparse tuning with low memory usage.
To achieve low memory usage, SNELL decomposes the tunable matrix for sparsification into two learnable low-rank matrices.
A competition-based sparsification mechanism is further proposed to avoid the storage of tunable weight indexes.
arXiv Detail & Related papers (2024-11-04T04:58:20Z) - A Stochastic Approach to Bi-Level Optimization for Hyperparameter Optimization and Meta Learning [74.80956524812714]
We tackle the general differentiable meta learning problem that is ubiquitous in modern deep learning.
These problems are often formalized as Bi-Level optimizations (BLO)
We introduce a novel perspective by turning a given BLO problem into a ii optimization, where the inner loss function becomes a smooth distribution, and the outer loss becomes an expected loss over the inner distribution.
arXiv Detail & Related papers (2024-10-14T12:10:06Z) - Memory-Efficient Optimization with Factorized Hamiltonian Descent [11.01832755213396]
We introduce a novel adaptive, H-Fac, which incorporates a memory-efficient factorization approach to address this challenge.
By employing a rank-1 parameterization for both momentum and scaling parameter estimators, H-Fac reduces memory costs to a sublinear level.
We develop our algorithms based on principles derived from Hamiltonian dynamics, providing robust theoretical underpinnings in optimization dynamics and convergence guarantees.
arXiv Detail & Related papers (2024-06-14T12:05:17Z) - Compute Better Spent: Replacing Dense Layers with Structured Matrices [77.61728033234233]
We identify more efficient alternatives to dense matrices, as exemplified by the success of convolutional networks in the image domain.
We show that different structures often require drastically different initialization scales and learning rates, which are crucial to performance.
We propose a novel matrix family containing Monarch matrices, the Block-Train, which we show performs better than dense for the same compute on multiple tasks.
arXiv Detail & Related papers (2024-06-10T13:25:43Z) - AdaLomo: Low-memory Optimization with Adaptive Learning Rate [59.64965955386855]
We introduce low-memory optimization with adaptive learning rate (AdaLomo) for large language models.
AdaLomo results on par with AdamW, while significantly reducing memory requirements, thereby lowering the hardware barrier to training large language models.
arXiv Detail & Related papers (2023-10-16T09:04:28Z) - CAME: Confidence-guided Adaptive Memory Efficient Optimization [20.009302737137787]
Adaptive gradient methods have demonstrated excellent performance in the training of large language models.
The need for maintaining second-moment estimates requires maintaining a high cost of extra memory overheads.
Several memory-efficients have been proposed to obtain a drastic reduction in auxiliary memory usage, but with a performance penalty.
We propose CAME to simultaneously achieve two goals: fast convergence as in traditional adaptive methods, and low memory usage as in memory-efficient methods.
arXiv Detail & Related papers (2023-07-05T06:05:36Z) - Efficient Learning of Generative Models via Finite-Difference Score
Matching [111.55998083406134]
We present a generic strategy to efficiently approximate any-order directional derivative with finite difference.
Our approximation only involves function evaluations, which can be executed in parallel, and no gradient computations.
arXiv Detail & Related papers (2020-07-07T10:05:01Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.