paper-with-me

홈 › Papers

Train Flat, Then Compress: Sharpness-Aware Minimization Learns More Compressible Models

2022-05-25 · Clara Na, Sanket Vaibhav Mehta, Emma Strubell

Model compression by way of parameter pruning, quantization, or distillation has recently gained popularity as an approach for reducing the computational requirements of modern deep neural network models for NLP. Inspired by prior works suggesting a connection between simpler, more generalizable models and those that lie within wider loss basins, we hypothesize that optimizing for flat minima should lead to simpler parameterizations and thus more compressible models. We propose to combine sharpness-aware minimization (SAM) with various task-specific model compression methods, including iterative magnitude pruning (IMP), structured pruning with a distillation objective, and post-training dynamic quantization. Empirically, we show that optimizing for flatter minima consistently leads to greater compressibility of parameters compared to vanilla Adam when fine-tuning BERT models, with little to no loss in accuracy on the GLUE text classification and SQuAD question answering benchmarks. Moreover, SAM finds superior winning tickets during IMP that 1) are amenable to vanilla Adam optimization, and 2) transfer more effectively across tasks.

📄 PDF Abstract BibTeX arXiv:2205.12694

Code (0)

등록된 구현이 없습니다.

Tasks

Model CompressionQuantizationQuestion Answeringtext-classificationText Classification

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 설명 없음
Pruning 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Weight Decay 설명 없음
Linear Warmup With Linear Decay Linear Warmup With Linear Decay is a learning rate schedule in which we increase the learning rate linearly for $n$ updates and then linearly decay afterwards.
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…

Similar Papers 제목 키워드 기반

Towards Compact and Robust DNNs via Compression-aware Sharpness Minimization

2026-01-28 · Jialuo He, Huangxun Chen arxiv

Sharpness-Aware Minimization (SAM) has recently emerged as an effective technique for improving DNN robustness to input variations. However, its interplay with the compactness requirements of on-device DNN deployments re…

A simple connection from loss flatness to compressed neural representations

2023-10-03 · Shirui Chen, Stefano Recanatesi, Eric Shea-Brown

Sharpness, a geometric measure in the parameter space that reflects the flatness of the loss landscape, has long been studied for its potential connections to neural network behavior. While sharpness is often associated …

Sensitivity

SADDLe: Sharpness-Aware Decentralized Deep Learning with Heterogeneous Data

2024-05-22 · Sakshi Choudhary, Sai Aparna Aketi, Kaushik Roy

Decentralized training enables learning with distributed datasets generated at different locations without relying on a central server. In realistic scenarios, the data distribution across these sparsely connected learni…

Deep Learning

Flat Seeking Bayesian Neural Networks

2023-02-06 · NeurIPS 2023 11

Bayesian Neural Networks (BNNs) provide a probabilistic interpretation for deep learning models by imposing a prior distribution over model parameters and inferring a posterior distribution based on observed data. The mo…

Deep LearningVariational Inference

Flat Minima and Generalization: Insights from Stochastic Convex Optimization

2025-11-05 · Matan Schliserman, Shira Vansover-Hager, Tomer Koren arxiv

Understanding the generalization behavior of learning algorithms is a central goal of learning theory. A recently emerging explanation is that learning algorithms are successful in practice because they converge to flat …