Topics

When features vastly outnumber samples, PLATO harnesses hidden knowledge graphs to regularize deep learning — boosting accuracy by over 10% in genomics and drug discovery.

Traditional graph models treat all nodes identically, erasing vital context. Learn how Het-node2vec utilizes type-aware random walks to capture the true meaning within complex networks.

Discover how lightweight graph models predict online purchases up to 180 times faster than heavy deep learning algorithms, outperforming complex neural networks with simple efficiency.

MAP, a knowledge-driven AI from Shanghai Jiao Tong, predicts single-cell responses for unprofiled drugs, prioritizing four of five approved cancer drugs in zero-shot tests.

NeuroMatch teaches AI to recognize hidden network patterns instantly, replacing slow graph searches with geometric intuition that scales across biology, chemistry, and knowledge systems.