\texttt{Range-Arithmetic}: Verifiable Deep Learning Inference on an Untrusted Party

TL;DR

Range-Arithmetic introduces a verifiable deep learning inference framework that efficiently transforms non-arithmetic operations into arithmetic steps, enabling resource-efficient verification suitable for decentralized systems.

Contribution

It presents a novel method to verify DNN inference by converting non-arithmetic operations into arithmetic steps using sum-check protocols and range proofs, avoiding complex encodings.

Findings

Reduces verification computational cost

Maintains performance comparable to existing methods

Lowers communication overhead in verification process

Abstract

Verifiable computing (VC) has gained prominence in decentralized machine learning systems, where resource-intensive tasks like deep neural network (DNN) inference are offloaded to external participants due to blockchain limitations. This creates a need to verify the correctness of outsourced computations without re-execution. We propose \texttt{Range-Arithmetic}, a novel framework for efficient and verifiable DNN inference that transforms non-arithmetic operations, such as rounding after fixed-point matrix multiplication and ReLU, into arithmetic steps verifiable using sum-check protocols and concatenated range proofs. Our approach avoids the complexity of Boolean encoding, high-degree polynomials, and large lookup tables while remaining compatible with finite-field-based proof systems. Experimental results show that our method not only matches the performance of existing approaches, but also reduces the computational cost of verifying the results, the computational effort required from the untrusted party performing the DNN inference, and the communication overhead between the two sides.