Pith. sign in

REVIEW

ModelObfuscator: Obfuscating Model Information to Protect Deployed ML-based Systems

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2306.06112 v3 pith:IU6PIQSJ submitted 2023-06-01 cs.CR

ModelObfuscator: Obfuscating Model Information to Protect Deployed ML-based Systems

classification cs.CR
keywords modelmodelson-deviceobfuscationinformationmodelobfuscatorappsattackers
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

More and more edge devices and mobile apps are leveraging deep learning (DL) capabilities. Deploying such models on devices -- referred to as on-device models -- rather than as remote cloud-hosted services, has gained popularity because it avoids transmitting user data off of the device and achieves high response time. However, on-device models can be easily attacked, as they can be accessed by unpacking corresponding apps and the model is fully exposed to attackers. Recent studies show that attackers can easily generate white-box-like attacks for an on-device model or even inverse its training data. To protect on-device models from white-box attacks, we propose a novel technique called model obfuscation. Specifically, model obfuscation hides and obfuscates the key information -- structure, parameters and attributes -- of models by renaming, parameter encapsulation, neural structure obfuscation obfuscation, shortcut injection, and extra layer injection. We have developed a prototype tool ModelObfuscator to automatically obfuscate on-device TFLite models. Our experiments show that this proposed approach can dramatically improve model security by significantly increasing the difficulty of parsing models inner information, without increasing the latency of DL models. Our proposed on-device model obfuscation has the potential to be a fundamental technique for on-device model deployment. Our prototype tool is publicly available at: https://github.com/zhoumingyi/ModelObfuscator.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.