Multiple Instance Learning with Bag Dissimilarities
This addresses the challenge of learning from ambiguous instance labels in MIL, offering a generalizable method that avoids dataset-specific assumptions, though it is incremental as it builds on existing dissimilarity-based approaches.
The paper tackles the problem of multiple instance learning (MIL) where instance labels are ambiguous, by proposing a method that represents bags as vectors of dissimilarities to other bags, treating them as features. The results show this approach is computationally inexpensive and competitive with state-of-the-art algorithms across various MIL datasets.
Multiple instance learning (MIL) is concerned with learning from sets (bags) of objects (instances), where the individual instance labels are ambiguous. In this setting, supervised learning cannot be applied directly. Often, specialized MIL methods learn by making additional assumptions about the relationship of the bag labels and instance labels. Such assumptions may fit a particular dataset, but do not generalize to the whole range of MIL problems. Other MIL methods shift the focus of assumptions from the labels to the overall (dis)similarity of bags, and therefore learn from bags directly. We propose to represent each bag by a vector of its dissimilarities to other bags in the training set, and treat these dissimilarities as a feature representation. We show several alternatives to define a dissimilarity between bags and discuss which definitions are more suitable for particular MIL problems. The experimental results show that the proposed approach is computationally inexpensive, yet very competitive with state-of-the-art algorithms on a wide range of MIL datasets.