Showing cs.LGShow all
2 papers · 1 filter
cs.LG2025
On the Mathematical Impossibility of Safe Universal Approximators
Jasper Yao
We establish fundamental mathematical limits on universal approximation theorem (UAT) system alignment by proving that catastrophic failures are an inescapable feature of any usefu…
cs.LG2025
AlgoSelect: Universal Algorithm Selection via the Comb Operator
Jasper Yao
We introduce AlgoSelect, a principled framework for learning optimal algorithm selection from data, centered around the novel Comb Operator. Given a set of algorithms and a feature…