A new paper, “Polynomial Time Cryptanalytic Extraction of Neural Network Models,” by Adi Shamir and others, uses ideas from differential cryptanalysis to extract the weights inside a neural network using specific queries and their results. This is much more theoretical than practical, but it’s a really interesting result.
Abstract:
Billions of dollars and countless GPU hours are currently spent on training Deep Neural Networks (DNNs) for a variety of tasks. Thus, it is essential to determine the difficulty of extracting all the parameters of such neural networks when given access to their black-box implementations. Many versions of this problem have been studied over the last 30 years, and the best current attack on ReLU-based deep neural networks was presented at Crypto’20 by Carlini, Jagielski, and Mironov. It resembles a differential chosen plaintext attack on a cryptosystem, which has a secret key embedded in its black-box implementation and requires a polynomial number of queries but an exponential amount of time (as a function of the number of neurons). In this paper, we improve this attack by developing several new techniques that enable us to extract with arbitrarily high precision all the real-valued parameters of a ReLU-based DNN using a polynomial number of queries and a polynomial amount of time. We demonstrate its practical efficiency by applying it to a full-sized neural network for classifying the CIFAR10 dataset, which has 3072 inputs, 8 hidden layers with 256 neurons each, and about 1.2 million neuronal parameters. An attack following the approach by Carlini et al. requires an exhaustive search over 2^256 possibilities. Our attack replaces this with our new techniques, which require only 30 minutes on a 256-core computer.
More Stories
Cloudflare Introduces E2E Post-Quantum Cryptography Protections
Cloudflare introduces E2E post-quantum cryptography, enhancing security against quantum threats Read More
UK’s Online Safety Act: Ofcom Can Now Issue Sanctions
From March 17, Ofcom will enforce rules requiring tech platforms operating in the UK to remove illegal content, including child...
Researchers Confirm BlackLock as Eldorado Rebrand
DarkAtlas researchers have uncovered a direct link between BlackLock and the Eldorado ransomware group, confirming a rebranded identity of the...
Improvements in Brute Force Attacks
New paper: “GPU Assisted Brute Force Cryptanalysis of GPRS, GSM, RFID, and TETRA: Brute Force Cryptanalysis of KASUMI, SPECK, and...
US Legislators Demand Transparency in Apple’s UK Backdoor Court Fight
A bipartisan delegation of US Congresspeople and Senators has asked the hearing between the UK government and Apple to be...
£1M Lost as UK Social Media and Email Account Hacks Skyrocket
Action Fraud reported a spike in social media and email account hacks in 2024, resulting in losses of nearly £1m...