Topological Inductive Bias fosters Multiple Instance Learning in Data-Scarce Scenarios

Salome Kazeminia, Carsten Marr, Bastian Rieck

Abstract: Multiple instance learning (MIL) is a framework for weakly supervised classification, where labels are assigned to sets of instances, i.e., bags, rather than to individual data points. This paradigm has proven effective in tasks where fine-grained annotations are unavailable or costly to obtain. However, the effectiveness of MIL drops sharply when training data are scarce, such as for rare disease classification. To address this challenge, we propose incorporating topological inductive biases into the data representation space within the MIL framework. This bias introduces a topology-preserving constraint that encourages the instance encoder to maintain the topological structure of the instance distribution within each bag when mapping them to MIL latent space. As a result, our Topology Guided MIL (TG-MIL) method enhances the performance and generalizability of MIL classifiers across different aggregation functions, especially under scarce-data regimes. Our evaluations show average performance improvements of 15.3% for synthetic MIL datasets, 2.8% for MIL benchmarks, and 5.5% for rare anemia classification compared to current state-of-the-art MIL models, where only 17–120 samples per class are available. We make our code publicly available at https://github.com/SalomeKaze/TGMIL.

Submission Type: Long submission (more than 12 pages of main content)

Changes Since Last Submission:

For the camera-ready revision, we have:
- Unanonymized the author list.
- Replaced the anonymized repository with the original GitHub repository URL.
- Performed a final round of manuscript polishing.
- Moved the Instance Learnability subsection to the Method section.
- Relocated the Broader Clinical Impact Statement to its designated section in the template.
- Added an Acknowledgements section.
No technical content, experimental results, model definitions, or mathematical formulations were modified. The core methodology and empirical findings remain unchanged.

Code: https://github.com/SalomeKaze/TGMIL

Assigned Action Editor: ~Adin_Ramirez_Rivera1

Submission Number: 6480