Non-Transferable Learning: A New Approach for Model Verification and Authorization

by   Lixu Wang, et al.

As Artificial Intelligence as a Service gains popularity, protecting well-trained models as intellectual property is becoming increasingly important. Generally speaking, there are two common protection methods: ownership verification and usage authorization. In this paper, we propose Non-Transferable Learning (NTL), a novel approach that captures the exclusive data representation in the learned model and restricts the model generalization ability to certain domains. This approach provides effective solutions to both model verification and authorization. For ownership verification, watermarking techniques are commonly used but are often vulnerable to sophisticated watermark removal methods. Our NTL-based model verification approach instead provides robust resistance to state-of-the-art watermark removal methods, as shown in extensive experiments for four of such methods over the digits, CIFAR10 STL10, and VisDA datasets. For usage authorization, prior solutions focus on authorizing specific users to use the model, but authorized users can still apply the model to any data without restriction. Our NTL-based authorization approach instead provides data-centric usage protection by significantly degrading the performance of usage on unauthorized data. Its effectiveness is also shown through experiments on a variety of datasets.


page 2

page 8

page 9

page 17

page 18

page 19

page 20


PromptCARE: Prompt Copyright Protection by Watermark Injection and Verification

Large language models (LLMs) have witnessed a meteoric rise in popularit...

OVLA: Neural Network Ownership Verification using Latent Watermarks

Ownership verification for neural networks is important for protecting t...

Towards Generalizable Data Protection With Transferable Unlearnable Examples

Artificial Intelligence (AI) is making a profound impact in almost every...

Domain Agnostic Few-shot Learning for Speaker Verification

Deep learning models for verification systems often fail to generalize t...

An Embarrassingly Simple Approach for Intellectual Property Rights Protection on Recurrent Neural Networks

Capitalise on deep learning models, offering Natural Language Processing...

A Unified Approach to Kinship Verification

In this work, we propose a deep learning-based approach for kin verifica...

A new approach (extra vertex) and generalization of Shoelace Algorithm usage in convex polygon (Point-in-Polygon)

In this paper we aim to bring new approach into usage of Shoelace Algori...

Please sign up or login with your details

Forgot password? Click here to reset