Attention Awareness Multiple Instance Neural Network
This work addresses a domain-specific issue in pattern recognition with weakly annotated data, offering an incremental improvement over existing MIL methods.
The paper tackles the problem of inflexible MIL pooling operators and inaccurate bag-level representations in multiple instance learning by proposing an attention awareness multiple instance neural network framework, which outperforms state-of-the-art MIL methods in experiments on pattern recognition tasks.
Multiple instance learning is qualified for many pattern recognition tasks with weakly annotated data. The combination of artificial neural network and multiple instance learning offers an end-to-end solution and has been widely utilized. However, challenges remain in two-folds. Firstly, current MIL pooling operators are usually pre-defined and lack flexibility to mine key instances. Secondly, in current solutions, the bag-level representation can be inaccurate or inaccessible. To this end, we propose an attention awareness multiple instance neural network framework in this paper. It consists of an instance-level classifier, a trainable MIL pooling operator based on spatial attention and a bag-level classification layer. Exhaustive experiments on a series of pattern recognition tasks demonstrate that our framework outperforms many state-of-the-art MIL methods and validates the effectiveness of our proposed attention MIL pooling operators.