A Note on Non-Composability of Layerwise Approximate Verification for Neural Inference

AI 总结

• A natural and informal approach to verifiable (or zero-knowledge) ML inference over floating-point data is: ``prove that each layer was com…
• This short note gives a simple counterexample showing that this inference is false in general: for any neural network, we can construct a f…

内容摘录

A natural and informal approach to verifiable (or zero-knowledge) ML inference over floating-point data is: ``prove that each layer was computed correctly up to tolerance $δ$; therefore the final output is a reasonable inference result''. This short note gives a simple counterexample showing that this inference is false in general: for any neural network, we can construct a functionally equivalent network for which adversarially chosen approximation-magnitude errors in individual layer computations suffice to steer the final output arbitrarily (within a prescribed bounded range).