Multi-task Joint Strategies of Self-supervised Representation Learning on Biomedical Networks for Drug Discovery
This work addresses the problem of optimizing multi-task SSL strategies for drug discovery, offering guidelines that could improve efficiency in biomedical research, though it appears incremental as it builds on existing SSL methods.
The paper tackled the challenge of effectively combining multiple self-supervised learning (SSL) models on biomedical networks for drug discovery by proposing MSSL2drug, which evaluated fifteen task combinations using a graph attention-based multi-task adversarial learning framework, finding that multimodal and local-global combinations achieved the best performance.
Self-supervised representation learning (SSL) on biomedical networks provides new opportunities for drug discovery. However, how to effectively combine multiple SSL models is still challenging and has been rarely explored. Therefore, we propose multi-task joint strategies of self-supervised representation learning on biomedical networks for drug discovery, named MSSL2drug. We design six basic SSL tasks inspired by various modality features including structures, semantics, and attributes in heterogeneous biomedical networks. Importantly, fifteen combinations of multiple tasks are evaluated by a graph attention-based multi-task adversarial learning framework in two drug discovery scenarios. The results suggest two important findings. (1) Combinations of multimodal tasks achieve the best performance compared to other multi-task joint models. (2) The local-global combination models yield higher performance than random two-task combinations when there are the same size of modalities. Therefore, we conjecture that the multimodal and local-global combination strategies can be treated as the guideline of multi-task SSL for drug discovery.