Multi-instance Learning as Downstream Task of Self-Supervised Learning-based Pre-trained Model

In deep multi-instance learning, the number of applicable instances depends on the data set. In histopathology images, deep learning multi-instance learners usually assume there are hundreds to thousands instances in a bag. However, when the number of instances in a bag increases to 256 in brain hematoma CT, learning becomes extremely difficult. In this paper, we address this drawback. To overcome this problem, we propose using a pre-trained model with self-supervised learning for the multi-instance learner as a downstream task. With this method, even when the original target task suffers from the spurious correlation problem, we show improvements of 5% to 13% in accuracy and 40% to 55% in the F1 measure for the hypodensity marker classification of brain hematoma CT.

@article{matsuishi2025_2505.21564,
  title={ Multi-instance Learning as Downstream Task of Self-Supervised Learning-based Pre-trained Model },
  author={ Koki Matsuishi and Tsuyoshi Okita },
  journal={arXiv preprint arXiv:2505.21564},
  year={ 2025 }
}