AINeutralarXiv – CS AI · 3h ago6/10
🧠
Guaranteed Optimal Compositional Explanations for Neurons
Researchers introduce the first framework for computing mathematically optimal compositional explanations of neural network neurons, replacing heuristic beam search methods that lack optimality guarantees. The work reveals that 10-40% of explanations previously generated by standard approaches are suboptimal when handling overlapping concepts, while proposing algorithms achieving comparable computational efficiency.