No abstract

Cognitive concept learning is to learn concepts from a given clue by simulating human thought processes including perception, attention and thinking. In recent years, it has attracted much attention from the communities of formal concept analysis, cognitive computing and granular computing. However, the classical cognitive concept learning approaches are not suitable for incomplete information. Motivated by this problem, this study mainly focuses on cognitive concept learning from incomplete information. Specifically, we put forward a pair of approximate cognitive operators to derive concepts from incomplete information. Then, we propose an approximate cognitive computing system to perform the transformation between granular concepts as incomplete information is updated periodically. Moreover, cognitive processes are simulated based on three types of similarities. Finally, numerical experiments are conducted to evaluate the proposed cognitive concept learning methods.

No abstract

Building models of natural language processing (NLP) is challenging in low-resource scenarios where only limited data are available. Optimization-based meta-learning algorithms achieve promising results in low-resource scenarios by adapting a well-generalized model initialization to handle new tasks. Nonetheless, these approaches suffer from the memorization overfitting issue, where the model tends to memorize the meta-training tasks while ignoring support sets when adapting to new tasks. To address this issue, we propose a memory imitation meta-learning (MemIML) method that enhances the model's reliance on support sets for task adaptation. Specifically, we introduce a task-specific memory module to store support set information and construct an imitation module to force query sets to imitate the behaviors of some representative supportset samples stored in the memory. A theoretical analysis is provided to prove the effectiveness of our method, and empirical results also demonstrate that our method outperforms competitive baselines on both text classification and generation tasks.

Based on the unique data from a popular P2P website in China, this paper investigates the borrowers’ herding behavior and its influencing factors, e.g., platform attribute and moderating events. We mainly find that: (i) there exists herding behavior at the platform level and that the loan amount, loan interest rate and loan duration are positively related to the magnitude of herding; (2) the borrowers’ herding behavior is increased by a platform’s market share and the number of participants; (3) regulatory events have negative effects on the herding behavior. Our findings have practical implications for P2P participants, platforms, and policymakers.