Human latent-state generalization through prototype learning with discriminative attention

People cannot access the latent causes giving rise to experience. How then do they approximate the high-dimensional feature space of the external world with lower-dimensional internal models that generalize to novel examples or contexts? Here, we developed and tested a theoretical framework that internally identifies states by feature regularity (i.e., prototype states) and selectively attends to features according to their informativeness for discriminating between likely states. To test theoretical predictions, we developed experimental tasks where human subjects first learn through reward-feedback internal models of latent states governing actions associated with multi-feature stimuli. We then analyzed subjects’ response patterns to novel examples and contexts. These combined theoretical and experimental results reveal that the human ability to generalize actions involves the formation of prototype states with flexible deployment of top-down attention to discriminative features. These cognitive strategies underlie the human ability to generalize learned latent states in high-dimensional environments.

Towards Open-Set Text Recognition via Label-to-Prototype Learning

<div><br></div><div>Scene text recognition is a popular topic and can benefit various tasks. Although many methods have been proposed for the close-set text recognition challenges, they cannot be directly applied to open-set scenarios, where the evaluation set contains novel characters not appearing in the training set. Conventional methods require collecting new data and retraining the model to handle these novel characters, which is an expensive and tedious process. In this paper, we propose a label-to-prototype learning framework to handle novel characters without retraining the model. In the proposed framework, novel characters are effectively mapped to their corresponding prototypes with a label-to-prototype learning module. This module is trained on characters with seen labels and can be easily generalized to novel characters. Additionally, feature-level rectification is conducted via topology-preserving transformation, resulting in better alignments between visual features and constructed prototypes while having a reasonably small impact on model speed. A lot of experiments show that our method achieves promising performance on a variety of zero-shot, close-set, and open-set text recognition datasets.</div>

Towards Open-Set Text Recognition via Label-to-Prototype Learning

<div><br></div><div>Scene text recognition is a popular topic and can benefit various tasks. Although many methods have been proposed for the close-set text recognition challenges, they cannot be directly applied to open-set scenarios, where the evaluation set contains novel characters not appearing in the training set. Conventional methods require collecting new data and retraining the model to handle these novel characters, which is an expensive and tedious process. In this paper, we propose a label-to-prototype learning framework to handle novel characters without retraining the model. In the proposed framework, novel characters are effectively mapped to their corresponding prototypes with a label-to-prototype learning module. This module is trained on characters with seen labels and can be easily generalized to novel characters. Additionally, feature-level rectification is conducted via topology-preserving transformation, resulting in better alignments between visual features and constructed prototypes while having a reasonably small impact on model speed. A lot of experiments show that our method achieves promising performance on a variety of zero-shot, close-set, and open-set text recognition datasets.</div>