A Vector Symbolic Approach to Multiple Instance Learning

2025/11/20 by Ehsan Ahmed Dhrubo, Dhrubo, Ehsan Ahmed, Mohammad Mahmudul Alam +7
Computer Science · #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning in Healthcare #Multimodal Machine Learning Applications

paper · pdf · doi:10.48550/arxiv.2511.16795

openalex publication_date 2025/11/20 · openalex created_date 2025/11/25 · openalex updated_date 2026/07/28

Abstract

Multiple Instance Learning (MIL) tasks impose a strict logical constraint: a bag is labeled positive if and only if at least one instance within it is positive. While this iff constraint aligns with many real-world applications, recent work has shown that most deep learning-based MIL approaches violate it, leading to inflated performance metrics and poor generalization. We propose a novel MIL framework based on Vector Symbolic Architectures (VSAs), which provide a differentiable mechanism for performing symbolic operations in high-dimensional space. Our method encodes the MIL assumption directly into the model's structure by representing instances and concepts as nearly orthogonal high-dimensional vectors and using algebraic operations to enforce the iff constraint during classification. To bridge the gap between raw data and VSA representations, we design a learned encoder that transforms input instances into VSA-compatible vectors while preserving key distributional properties. Our approach, which includes a VSA-driven MaxNetwork classifier, achieves state-of-the-art results for a valid MIL model on standard MIL benchmarks and medical imaging datasets, outperforming existing methods while maintaining strict adherence to the MIL formulation. This work offers a principled, interpretable, and effective alternative to existing MIL approaches that rely on learned heuristics.

Citations

• Linearithmic Clean-up for Vector-Symbolic Key-Value Memory with Kroneker Rotation Products
• A Walsh Hadamard Derived Linear Vector Symbolic Architecture
• xMIL: Insightful Explanations for Multiple Instance Learning in Histopathology
• Assembling Modular, Hierarchical Cognitive Map Learners with Hyperdimensional Computing
• Holographic Global Convolutional Networks for Long-Range Prediction Tasks in Malware Detection
• Towards Generalization in Subitizing with Neuro-Symbolic Loss using Holographic Reduced Representations
• Reproducibility in Multiple Instance Learning: A Case For Algorithmic Unit Tests
• Recasting Self-Attention with Holographic Reduced Representations
• Deploying Convolutional Networks on Untrusted Platforms Using 2D Holographic Reduced Representations
• Multiple Instance Learning for Digital Pathology: A Review on the State-of-the-Art, Limitations & Future Potential
• GraphHD: Efficient graph classification using hyperdimensional computing
• A Neuro-vector-symbolic Architecture for Solving Raven's Progressive Matrices
• Learning with Holographic Reduced Representations
• TransMIL: Transformer based Correlated Multiple Instance Learning for Whole Slide Image Classification
• Dual-stream Multiple Instance Learning Network for Whole Slide Image Classification with Self-supervised Contrastive Learning
• Bringing UMAP Closer to the Speed of Light with GPU Acceleration
• A Metric Learning Reality Check
• ReZero is All You Need: Fast Convergence at Large Depth
• Multiple instance learning with graph neural networks
• Vector-Derived Transformation Binding: An Improved Binding Operation for Deep Symbol-Like Processing in Neural Networks
• Attention-based Deep Multiple Instance Learning
• UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction
• Associative Long Short-Term Memory
• Multi-Instance Learning by Treating Instances As Non-I.I.D. Samples
• Solving the multiple instance problem with axis-parallel rectangles

Related

• Using Neural Network Formalism to Solve Multiple-Instance Problems
• ProMIL: Probabilistic Multiple Instance Learning for Medical Imaging
• Attention-based Deep Multiple Instance Learning
• Rethinking Multiple Instance Learning: Developing an Instance-Level Classifier via Weakly-Supervised Self-Training
• mil-benchmarks: Standardized Evaluation of Deep Multiple-Instance Learning Techniques