Multi-Task Self-Training for Learning General Representations

TL;DR

This paper introduces MuST, a multi-task self-training approach that leverages specialized teacher models to create a unified student model capable of performing well across diverse vision tasks, using both labeled and unlabeled data.

Contribution

The paper presents a novel multi-task self-training framework that effectively combines knowledge from multiple specialized teachers to train a generalist model for various vision tasks.

Findings

MuST outperforms specialized models on multiple vision tasks.

The approach scales well with large unlabeled datasets.

MuST improves upon existing checkpoints trained with billions of examples.

Abstract

Despite the fast progress in training specialized models for various tasks, learning a single general model that works well for many tasks is still challenging for computer vision. Here we introduce multi-task self-training (MuST), which harnesses the knowledge in independent specialized teacher models (e.g., ImageNet model on classification) to train a single general student model. Our approach has three steps. First, we train specialized teachers independently on labeled datasets. We then use the specialized teachers to label an unlabeled dataset to create a multi-task pseudo labeled dataset. Finally, the dataset, which now contains pseudo labels from teacher models trained on different datasets/tasks, is then used to train a student model with multi-task learning. We evaluate the feature representations of the student model on 6 vision tasks including image recognition (classification, detection, segmentation)and 3D geometry estimation (depth and surface normal estimation). MuST is scalable with unlabeled or partially labeled datasets and outperforms both specialized supervised models and self-supervised models when training on large scale datasets. Lastly, we show MuST can improve upon already strong checkpoints trained with billions of examples. The results suggest self-training is a promising direction to aggregate labeled and unlabeled training data for learning general feature representations.