首页 > AI前沿 > Rank Portability Does Not Imply Feasibility Portability: Target-Specific Evaluation of Joint Hardware Constraints

Rank Portability Does Not Imply Feasibility Portability: Target-Specific Evaluation of Joint Hardware Constraints

arXiv机器学习 2026-08-20 11:14 4 阅读 查看原文

Cross-device hardware evaluation often assumes that if architecture rankings transfer across devices, a proxy device can support target-side model selection.

We stress-test this assumption for joint latency-energy feasibility across two public architecture families.

On NAS-Bench-201

Cross-device rank correlations are moderate, while target-comparable feasible-set overlap remains incomplete.

A faithful AdaProxy diagnostic substantially improves latency ranking, showing that the observed boundary failures are not simply due to weak adaptation.

Exact finite-sample split-conformal analysis

Exact finite-sample split-conformal analysis also exposes an evidence bottleneck: a finite one-sided 90% threshold requires at least nine calibration observations.

We then replicate the phenomenon on 10,000 GPT architectures across 13 HW-GPT-Bench devices.

Relative to an RTX3080 proxy

Relative to an RTX3080 proxy, target latency SRCC ranges from 0.951 to 0.996, yet proxy-reuse violation risk ranges from 33.3% to 100% under matched joint constraints.

These results show

These results show that rank portability, feasibility portability, and target-specific decision support are distinct evaluation objects.

Cross-device evaluations should therefore report which target environments actually support the operating point being claimed.