ITP-quantumseminare December 2020

itp-quantumseminare@lists.phys.ethz.ch

2 participants
4 discussions

tomorrow's QIT Seminar, 2pm zoom
by Joe Renes 21 Dec '20

21 Dec '20

Hi all, Tomorrow at 2pm Jan Li will tell us about his master thesis on "Device-independent Quantum Key distribution and Nonlocality Distillation", which he did with Dr. Koon Tong Goh and Prof. Dr. Charles Ci Wen Lim at CQT and NUS in Singapore. See below for the abstract. The zoom link is https://ethz.zoom.us/j/362994444. Best, Joe Abstract: In this project, we have investigated the use of local wirings as described in Short et al. [table 1 https://arxiv.org/abs/quant-ph/0508120] to revive zero keyrates of DIQKD set-ups to positive keyrates. In this project, we first considered the correlations generated by a two qubit set-up. Due to the high computational cost of considering all possible wirings, we checked for a subclass of extremal wirings if there is revival. For this model, we found that some revival was possible for some wirings and that revival could be pushed further by wiring multiple times. Next, we moved to a more realistic model, the all-photonics system as is considered in Tsujimoto et al. [ https://journals.aps.org/pra/abstract/10.1103/PhysRevA.98.063842]. For this model however, we were not able to find any improvements and further research is needed.

1 0

QIT Seminar 15 December, 2pm
by Joe Renes 14 Dec '20

14 Dec '20

Hi all, Tomorrow Amira Abbas from IBM Zürich and the University of Durban will tell us about "The power of quantum neural networks". The full abstract is below. Join us at 2pm on zoom, at https://ethz.zoom.us/j/362994444. Best, Joe Abstract: Fault-tolerant quantum computers offer the promise of dramatically improving machine learning through speed-ups in computation or improved model scalability. In the near-term, however, the benefits of quantum machine learning are not so clear. Understanding expressibility and trainability of quantum models–-and quantum neural networks in particular–-requires further investigation. In this work, we use tools from information geometry to define a notion of expressibility for quantum and classical models. The effective dimension, which depends on the Fisher information, is used to prove a novel generalisation bound and establish a robust measure of expressibility. We show that quantum neural networks are able to achieve a significantly better effective dimension than comparable classical neural networks. To then assess the trainability of quantum models, we connect the Fisher information spectrum to barren plateaus, the problem of vanishing gradients. Importantly, certain quantum neural networks can show resilience to this phenomenon and train faster than classical models due to their favourable optimisation landscapes, captured by a more evenly spread Fisher information spectrum. Our work is the first to demonstrate that well-designed quantum neural networks offer an advantage over classical neural networks through a higher effective dimension and faster training ability, which we verify on real quantum hardware.

1 0

Tuesday's QIT Seminar
by Joe Renes 07 Dec '20

07 Dec '20

Hi all, Tuesday at 2pm Alexander Miessen will present his master thesis on "Variational Quantum Algorithms for Real Space Wave Packet Dynamics", which he carried out with Ivano Tavernelli at IBM Research. See below for the abstract. The zoom link is https://ethz.zoom.us/j/362994444. Note that at 3pm Andru Gheorghiu will deliver the ITS Fellows' Seminar on "Quantum computation: hardness, verification and protocols for near-term devices". See https://eth-its.ethz.ch/activities/its-fellows--seminar/Alexandru-Gheorghiu… for more details. The zoom link is https://ethz.zoom.us/j/99922681517. Best, Joe Abstract: Variational quantum algorithms (VQAs) have become the new paradigm in applied quantum computing, enabling calculations on current-day quantum hardware that would be far out of reach otherwise. The triumph of hybrid quantum-classical schemes over purely-quantum methods has led to numerous further-developments and applications to stationary problems in recent years, yet the most widely used method for quantum dynamics simulation remains the Trotterization of the unitary time-evolution operator. However, it relies on deep quantum circuits and is therefore hampered by the strong limitations of available quantum technology which prohibits its use for practical tasks in the near future. The development of a variational time-evolution (VTE) algorithm was therefore an important and necessary s tep for quantum dynamics to gain more foothold in present-day quantum computing. Nonetheless, VTE faces the difficulty of finding an ansatz for expressing the wavefunction which is able to maintain satisfactory accuracy throughout an evolution without diverging from the true state trajectory. While heuristic ansätze enjoy popularity, findings suggest that they require exponentially many parameters to offer enough flexibility, jeopardizing any potential quantum advantage. One hypothesized way out of this dilemma is the use of a physically motivated ansatz, the quantum circuit analog of an educated guess, which samples from a reduced variational space by taking into consideration model-specific symmetries. In this work, we aim to explore this possibility by studying the dynamics of molecular vibrations as well as a spin-boson model. We analyze existing physically motivated ansätze, developed for stationary VQAs, and propose modifications thereof to make them better suited for VTE. Subsequently, they are tested in the simulation of various system's dynamics. Our results indicate that it is indeed possible to find a compact yet accurate variational ansatz by leveraging symmetries of the system under study, requiring far less than exponentially many parameters.

1 0

Bitcoin auf dem Weg zum Rekord ↑
by Bitcoin System 02 Dec '20

02 Dec '20

Bitcoin System Erfahrungen – Funktioniert Bitcoin System wirklich? Laufend gibt es neue Plattformen, welche sich auf den Handel mit Kryptowährung aller Art spezialisiert haben. Doch wie sollen Kunden und Anleger noch herausfinden, welche wirklich seriös sind und bei welchen sich die Anmeldung überhaupt lohnt? Derzeit gibt es Tausende verschiedene Währungen – auch hier ist es schwierig, als unerfahrener Anleger den Überblick zu wahren. Umso wichtiger ist ein Anbieter, der das Trading durch eine übersichtliche Benutzeroberfläche und eine leichte Bedienung einfacher macht. Generell fragen sich sehr viele Menschen gerade an Anfang, ob die Nutzung eines Bots in ihrem individuellen Fall überhaupt Sinn ergibt. Viele entscheiden sich schlussendlich dafür, weil es sich um eine Möglichkeit handelt, welche auch Laien direkt effektiv nutzen können. Bei Bitcoin System handelt es sich nur um einen von vielen Bots, doch dieser gilt als zuverlässig und einsteigerfreundlich. Unser Fazit vorab: Bitcoin System hat uns Überzeugt und ist kein Betrug! JETZT KOSTENLOSES BITCOIN SYSTEM DEMOKONTO ERÖFFNEN >>>

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

ITP-quantumseminare December 2020