Meet Hybrid Quantum Machine Learning (HQML) for Cybersecurity
And, so, we stand on the horizon of one in every of the best advancements in science and technology: the . For the primary time, we’ll give you the option to compute with light particles and move away from our traditional Von Neumann architecture. In a decade or so, perhaps, our silicon-based chips will feel just like the — clunky, inefficient, and slow. And, the entire concept of pushing electrons down npn transistor junctions for computation may feel like we’re racing on a skateboard against a Jumbo Jet at top speed.
The quantum computer, though, will disrupt virtually every thing to do with trust and security but will bring great advancements in computation. While not every problem will give you the option to be processed by quantum computers, there are some which will be significantly enhanced. So, let’s take a look at the risks of the quantum computer, and description how they will be utilized in cybersecurity evaluation. For this, I’ll outline our recent paper which advances the realm.
The threats? … breaking the Web
With the rise of quantum computers, we’ll see many risks and opportunities. On the danger side, we face an untrusted digital world, and where of our existing public key encryption methods will be cracked by quantum computers. This includes RSA, ECC, and Discrete Logs, and for key exchange (ECDH), digital signatures (RSA, ECDSA, and EdDSA), and public key encryption (RSA and ElGamal).
Luckily, there are numerous (PQC) standards which are being developed, and which can work alongside our existing public key methods — a hybrid approach, before the eventual switch-off of RSA and ECC. These include CRYSTALS-Kyber (for key exchange and public key encryption) and CRYSTALS-Dilithium, SPHINCS+, and Falcon (for digital signatures). A lot of the proposed methods use lattice cryptography, and which has been shown to be robust against quantum attacks. If you need to learn more, try:
The opportunities? … an entire recent world of processing
But, what concerning the opportunities? Well, quantum computers will bring a recent way of process, and use to interchange our binary digits. These approaches will allow us to develop recent approaches to existing problems.
In cybersecurity, our forthcoming Ph.D. student (Madjid Tehrani) and collaborators from Capgemini (Eldar Sultanow and Anja Jeschke), McGill University (Malik Amir), George Washington University (Raymond Chow), and Edinburgh Napier University (myself and Mouad Lemoudden) have just outlined a recent approach to merging quantum computers and machine learning so as to solve a well known cybersecurity problem [here]:
This implements hybrid machine learning methods on real quantum computers, with 100 data samples, and likewise with real-device-based simulations, with 5,000 data samples. It uses HQML algorithms applied to the detection of botnet-generated domains. This includes the features of the character length of the domain, the entropy of the domain name, and the fame of the domain name:
With domain name evaluation we regularly use entropy measurements — and where we measure the variability of the letters utilized in the domain name. Together with this, n-Gram detection analyses 1, 2, 3 and more character sequences and match them to their relative probability and pronounceability. For instance, “qq” is a highly improbable 2-Gram sequence, but “tr” is.
The advancements within the paper are a “stabilized quantum architecture that allows us to execute HQML algorithms on real quantum devices” and “the design of a recent type of hybrid quantum binary classification algorithms which are based on ”.
If you happen to have an interest, the Hoeffding tree was first defined in 1963 [2]:
and where recent work has considerably speeded-up the method:
The experiments used the Qiskit library with Aer quantum simulator, together with using three different real quantum devices (on the Microsoft Azure cloud): IonQ, Rigetti, and Quantinuum, and that is the primary time that these tools have been combined.
Conclusions
We consider the paper is a terrific advancement in applying hybrid quantum machine learning to the sphere of cybersecurity analytics. So, don’t let your organization stand blindfolded to the threats and opportunities of quantum computers — prepare!
References
[1] Madjid Tehrani, Eldar Sultanow, William J Buchanan, Malik Amir and Anja Jeschke, Raymond Chow and Mouad Lemoudden, Enabling Quantum Cybersecurity Analytics in Botnet Detection: Stable Architecture and Speed-up through Tree Algorithms, 2023, https://arxiv.org/abs/2306.13727.
[2] Hoeffding, W. (1994). Probability inequalities for sums of bounded random variables. The collected works of Wassily Hoeffding, 409–426.