Hi all,
tomorrow Gian Gentinetta will tell us about his master thesis at IBM, entitled ''The complexity of quantum support vector machines''. See below for the abstract. The talk will take place at 2pm in F31.1 or on Zoom: https://ethz.zoom.us/j/362994444.
Best,
Ladina
%%%%%
Abstract:
Quantum support vector machines employ quantum circuits to define the kernel function. It has been shown that this approach offers a provable exponential speedup compared to any known classical algorithm for certain data sets. The training of such models corresponds to solving a convex optimization problem either via its primal or dual formulation. Due to the probabilistic nature of quantum mechanics, the training algorithms are affected by statistical uncertainty, which has a major impact on their complexity. We show that the dual problem can be solved in O(M^4.67/ε^2) quantum circuit evaluations, where M denotes the size of the data set and ε the solution accuracy. We prove under an empirically motivated assumption that the kernelized primal problem can alternatively be solved in O(min{M^2/ε^6, 1/ε^10}) evaluations by employing a generalization of a known classical algorithm called Pegasos. Accompanying empirical results demonstrate these analytical complexities to be essentially tight. In addition, we investigate a variational approximation to quantum support vector machines and show that their heuristic training achieves considerably better scaling in our experiments.
Hi all,
Tomorrow Martin Sandfuchs will tell us about his project with Ramona, entitled ''Computing key rates for device-independent quantum key distribution''. See below for the abstract. The talk will take place at 2pm in HIT F31.1 or on Zoom: https://ethz.zoom.us/j/362994444.
Best,
Ladina
%%%%%
Abstract:
Quantum key distribution (QKD) aims to exploit the properties of quantum mechanics to produce a shared secret key between two distant users that is uncorrelated with the information held by an adversary. Device-independent QKD (DIQKD) tries to provide even stronger security guarantees by removing assumptions about how the devices operate. Unfortunately physical implementations of DIQKD pose significant experimental challenges and as a reaction a lot of theoretical work has gone into inventing improved protocols that can increase the tolerance to noise. A recent numerical method, developed by Peter Brown et al., has allowed for improved key rates in the asymptotic limit. In this presentation we demonstrate how this technique can be extended to the more realistic regime of finitely many key rounds. To do so we lower-bound the smooth min-entropy using the entropy accumulation theorem. We then build on previous DIQKD security proofs to propose a modified protocol which improves the threshold efficiency when performing DIQKD with lossy qubits.
Hi all,
Tomorrow Xavier Coiteux-Roy of USI Lugano will tell us about
"Quantum-resistant cryptography for Bennett-Maxwell demons". See below for
the abstract. The meeting will take place in HIT F13 and on zoom at the
usual address https://ethz.zoom.us/j/362994444.
Best,
Joe
%%%%%%
Title: Quantum-resistant cryptography for Bennett-Maxwell demons
Abstract: I will present how the second law of thermodynamics allows in
principle to achieve information-theoretically secure cryptographic
primitives such as secret-key establishment and oblivious transfer. While
idealized, our proposed protocols explore the limits of logically and
thermodynamically reversible computation and act as a focal point to
contrast the properties of information in different physical theories.