CSI: Contrastive Data Stratification for Interaction Prediction and its Application to Compound-Protein Interaction Prediction
This addresses a fundamental interaction prediction problem in domains like bioinformatics and recommendation systems, but appears incremental as it builds on existing contrastive learning techniques.
The paper tackles the problem of predicting interactions between objects (e.g., compound-protein) by using data stratification and contrastive learning to enhance representations, resulting in improved accuracy, though no concrete numbers are provided in the abstract.
Accurately predicting the likelihood of interaction between two objects (compound-protein sequence, user-item, author-paper, etc.) is a fundamental problem in Computer Science. Current deep-learning models rely on learning accurate representations of the interacting objects. Importantly, relationships between the interacting objects, or features of the interaction, offer an opportunity to partition the data to create multi-views of the interacting objects. The resulting congruent and non-congruent views can then be exploited via contrastive learning techniques to learn enhanced representations of the objects.