Intellectual Property Protection for Deep Learning Model and Dataset Intelligence

TL;DR

This paper provides a comprehensive survey of intellectual property protection methods for both deep learning models and datasets, addressing evaluation metrics, attack vectors, and challenges in distributed training environments.

Contribution

It uniquely covers IP protection strategies for both models and datasets, including evaluation metrics, attack analysis, and challenges specific to distributed training settings.

Findings

Summarizes performance evaluation metrics for IPP methods.

Analyzes existing IPP techniques from proactive and reactive perspectives.

Discusses challenges and attacks faced by deep IPP in various training environments.

Abstract

With the growing applications of Deep Learning (DL), especially recent spectacular achievements of Large Language Models (LLMs) such as ChatGPT and LLaMA, the commercial significance of these remarkable models has soared. However, acquiring well-trained models is costly and resource-intensive. It requires a considerable high-quality dataset, substantial investment in dedicated architecture design, expensive computational resources, and efforts to develop technical expertise. Consequently, safeguarding the Intellectual Property (IP) of well-trained models is attracting increasing attention. In contrast to existing surveys overwhelmingly focusing on model IPP mainly, this survey not only encompasses the protection on model level intelligence but also valuable dataset intelligence. Firstly, according to the requirements for effective IPP design, this work systematically summarizes the general and scheme-specific performance evaluation metrics. Secondly, from proactive IP infringement prevention and reactive IP ownership verification perspectives, it comprehensively investigates and analyzes the existing IPP methods for both dataset and model intelligence. Additionally, from the standpoint of training settings, it delves into the unique challenges that distributed settings pose to IPP compared to centralized settings. Furthermore, this work examines various attacks faced by deep IPP techniques. Finally, we outline prospects for promising future directions that may act as a guide for innovative research.