SpanKey: Dynamic Key Space Conditioning for Neural Network Access Control

TL;DR

SpanKey introduces a lightweight, key-conditioned gating mechanism for neural network inference that controls access via subspace keys, with analytical insights and experimental validation on CIFAR-10 and MNIST.

Contribution

The paper proposes a novel subspace key injection method for neural network gating, along with analytical diagnostics and experimental evaluations demonstrating its effectiveness.

Findings

Effective gating on CIFAR-10 ResNet-18

Analytical results on key absorption and energy split

A threat discussion clarifies security limitations

Abstract

SpanKey is a lightweight way to gate inference without encrypting weights or chasing leaderboard accuracy on gated inference. The idea is to condition activations on secret keys. A basis matrix $B$ defines a low-dimensional key subspace $Span(B)$; during training we sample coefficients $\alpha$ and form keys $k=\alpha^\top B$, then inject them into intermediate activations with additive or multiplicative maps and strength $\gamma$. Valid keys lie in $Span(B)$; invalid keys are sampled outside that subspace. We make three points. (i) Mechanism: subspace key injection and a multi-layer design space. (ii) Failure mode: key absorption, together with two analytical results (a Beta-energy split and margin-tail diagnostics), explains weak baseline separation in energy and margin terms -- these are not a security theorem. iii) Deny losses and experiments: Modes A--C and extensions, with CIFAR-10 ResNet-18 runs and MNIST ablations for Mode B. We summarize setup and first-order analysis, injectors, absorption, deny losses and ablations, a threat discussion that does not promise cryptography, and closing remarks on scale. Code: \texttt{https://github.com/mindmemory-ai/dksc}