Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Towards Stable and Storage-efficient Dataset Distillation: Matching Convexified Trajectory

About

The rapid evolution of deep learning and large language models has led to an exponential growth in the demand for training data, prompting the development of Dataset Distillation methods to address the challenges of managing large datasets. Among these, Matching Training Trajectories (MTT) has been a prominent approach, which replicates the training trajectory of an expert network on real data with a synthetic dataset. However, our investigation found that this method suffers from three significant limitations: 1. Instability of expert trajectory generated by Stochastic Gradient Descent (SGD); 2. Low convergence speed of the distillation process; 3. High storage consumption of the expert trajectory. To address these issues, we offer a new perspective on understanding the essence of Dataset Distillation and MTT through a simple transformation of the objective function, and introduce a novel method called Matching Convexified Trajectory (MCT), which aims to provide better guidance for the student trajectory. MCT leverages insights from the linearized dynamics of Neural Tangent Kernel methods to create a convex combination of expert trajectories, guiding the student network to converge rapidly and stably. This trajectory is not only easier to store, but also enables a continuous sampling strategy during distillation, ensuring thorough learning and fitting of the entire expert trajectory. Comprehensive experiments across three public datasets validate the superiority of MCT over traditional MTT methods.

Wenliang Zhong, Haoyu Tang, Qinghai Zheng, Mingzhu Xu, Yupeng Hu, Liqiang Nie• 2024

Related benchmarks

TaskDatasetResultRank
Image ClassificationCIFAR-10
Accuracy72.3
507
Image ClassificationCIFAR-100
Nominal Accuracy46.8
116
In-hospital mortality predictioneICU (test)--
78
Image ClassificationTiny-ImageNet
Validation Accuracy27.6
57
COVID-19 predictionOxford 1.7% prevalence (test)
AUROC0.882
25
COVID-19 predictionOxford 1.7% prevalence
AUROC0.882
25
COVID-19 predictionBirmingham 0.8% prevalence
AUROC88.6
25
COVID-19 predictionBirmingham NHS cohort 0.8% prevalence (test)
AUROC88.6
25
COVID-19 predictionPortsmouth NHS cohort 5.3% prevalence
AUROC0.893
25
Multi-label phenotypingMIMIC-III
Macro AUROC0.714
25
Showing 10 of 10 rows

Other info

Follow for update