UNIFORM: Unifying Knowledge from Large-scale and Diverse Pre-trained Models

TL;DR

UNIFORM is a novel framework that unifies knowledge from diverse pre-trained models into a single student model, overcoming heterogeneity and improving unsupervised object recognition, especially at large scale.

Contribution

It introduces a voting-based knowledge transfer method that works across different architectures and label spaces without strong assumptions, enabling scalable integration of many models.

Findings

Enhances unsupervised object recognition performance.

Scales effectively with over one hundred teacher models.

Outperforms existing knowledge transfer baselines.

Abstract

In the era of deep learning, the increasing number of pre-trained models available online presents a wealth of knowledge. These models, developed with diverse architectures and trained on varied datasets for different tasks, provide unique interpretations of the real world. Their collective consensus is likely universal and generalizable to unseen data. However, effectively harnessing this collective knowledge poses a fundamental challenge due to the heterogeneity of pre-trained models. Existing knowledge integration solutions typically rely on strong assumptions about training data distributions and network architectures, limiting them to learning only from specific types of models and resulting in data and/or inductive biases. In this work, we introduce a novel framework, namely UNIFORM, for knowledge transfer from a diverse set of off-the-shelf models into one student model without such constraints. Specifically, we propose a dedicated voting mechanism to capture the consensus of knowledge both at the logit level -- incorporating teacher models that are capable of predicting target classes of interest -- and at the feature level, utilizing visual representations learned on arbitrary label spaces. Extensive experiments demonstrate that UNIFORM effectively enhances unsupervised object recognition performance compared to strong knowledge transfer baselines. Notably, it exhibits remarkable scalability by benefiting from over one hundred teachers, while existing methods saturate at a much smaller scale.