Breaking Class Barriers: Efficient Dataset Distillation via Inter-Class Feature Compensator

TL;DR

This paper introduces INFER, a novel dataset distillation method that improves efficiency and effectiveness by leveraging inter-class feature compensation, surpassing traditional class-specific approaches and achieving state-of-the-art results.

Contribution

INFER employs a Universal Feature Compensator to facilitate inter-class feature integration, reducing label complexity and enhancing synthetic data quality in dataset distillation.

Findings

Outperforms SRe2L by 34.5% on ImageNet-1k at ipc=50 with ResNet18

Significantly reduces soft label size in synthetic datasets

Establishes new benchmarks in efficiency and effectiveness

Abstract

Dataset distillation has emerged as a technique aiming to condense informative features from large, natural datasets into a compact and synthetic form. While recent advancements have refined this technique, its performance is bottlenecked by the prevailing class-specific synthesis paradigm. Under this paradigm, synthetic data is optimized exclusively for a pre-assigned one-hot label, creating an implicit class barrier in feature condensation. This leads to inefficient utilization of the distillation budget and oversight of inter-class feature distributions, which ultimately limits the effectiveness and efficiency, as demonstrated in our analysis. To overcome these constraints, this paper presents the Inter-class Feature Compensator (INFER), an innovative distillation approach that transcends the class-specific data-label framework widely utilized in current dataset distillation methods. Specifically, INFER leverages a Universal Feature Compensator (UFC) to enhance feature integration across classes, enabling the generation of multiple additional synthetic instances from a single UFC input. This significantly improves the efficiency of the distillation budget. Moreover, INFER enriches inter-class interactions during the distillation, thereby enhancing the effectiveness and generalizability of the distilled data. By allowing for the linear interpolation of labels similar to those in the original dataset, INFER meticulously optimizes the synthetic data and dramatically reduces the size of soft labels in the synthetic dataset to almost zero, establishing a new benchmark for efficiency and effectiveness in dataset distillation. In practice, INFER demonstrates state-of-the-art performance across benchmark datasets. For instance, in the ipc = 50 setting on ImageNet-1k with the same compression level, it outperforms SRe2L by 34.5% using ResNet18.