Kwame Agyei-Baah

1 articles · 2 co-authors · latest: 2026-01-17
0 citations · h-index 0 · 1 papersData: Semantic Scholar

We have retold 1 papers by this author. most often about: Explainable AI, neural networks, evidential neural networks. 2 co-authors across our retellings.

Links in the knowledge graph
Fluid Dynamics2026-01-172 minarXiv:2601.11946express
The Secret of Transparency: How a Neural Network Learns from Water
Water in a stream bends around rocks, but the math of that motion hides in a black box: that's how ordinary neural networks work. They give an answer but don't say why. Engineers can't trust a prediction if they can't see its reasoning. Fluid dynamics now has a transparent model.