#QuantumMachineLearning

Single-shot quantum machine learning This work brings together notions of quantum machine learning with those of quantum metrology. In detail, #quantummachinelearning aims to improve learning methods through the use of quantum computers. If it is to ever realize its potential, many obstacles need to be overcome. A particularly pressing one arises at the prediction stage because the outputs of quantum learning models are inherently random. This creates an often considerable overhead, as many executions of a quantum learning model have to be aggregated to obtain an actual prediction. In this work, we analyze when quantum learning models can evade this issue and produce predictions in a near-deterministic way—paving the way to single-shot quantum machine learning. We give a rigorous definition of #singleshotness in quantum classifiers and show that the degree to which a quantum learning model is near deterministic is constrained by the distinguishability of the embedded quantum states used in the model. Opening the black box of the embedding, we show that if the embedding is realized by quantum circuits, a certain depth is necessary for single shotness to be even possible. We conclude by showing that quantum learning models cannot be single shot in a generic way and trainable at the same time. Warm thanks to @jj_xyz and Erik Recio Armengol.

In our reading group, @IMathYou2 beautifully explains topics in #quantummachinelearning.

An unconditional distribution learning advantage with shallow quantum circuits We present a #quantummachinelearning advantage of families of constant depth local quantum circuits over reasonably constrained log-log-depth classical circuits.

Learnable and unlearnable distributions in #quantummachinelearning. This is also the forth talk in four consecutive days, but now I am looking forward to slow discussions for a few days in wonderful Seefeld.

"Quantum Neural Networks, Computational Field Theory and Dynamics" #QuantumAI #QuantumMachineLearning #MachineLearning #ArtificialIntelligence #ComplexityScience #ChaosTheory #QuantumChaos #QuantumComplexity #EdgeOfChaos Full article at:

The exploitation of #symmetry is much at home in theoretical physics, specifically in quantum many-body theory. Here, we explore how symmetry can be made use of in variational #quantummachinelearning, where symmetries can also enter on the data level.

A collaboration of Andi Bauer of our team and that of Seth Lloyd shows that a #quantummachinelearning ansatz is too symmetric in the sense of then it can be classically efficiently simulated.

Are the output distributions of #quantumcircuits actually #learnable? We systematically look at this question, within two different oracle models, presenting a positive and a negative result.

As a positive result, show that in a more powerful oracle model, namely when directly given access to samples, the output distributions of local Clifford circuits are computationally efficiently PAC learnable by a classical learner.

On a higher level, this work discusses the connection between learnability and efficient simulatability, and contributes to a rigorous picture of notions of learning in the context of #quantumcomputing. #quantummachinelearning

I have completed the Quantum Machine Learning certification issued by @IBM . Credential: #QuantumComputing #QuantumMachineLearning #QML #IBMQuantum #VariationalQuantumCircuits #ArtificialIntelligence #Algorithms #SoftwareArchitecture #Qiskit #IBM

Hello, many worlds. We’re Cascade Quantum. Our goal is to make it effortless to create elegant quantum applications. Here's what we're building: #QuantumComputing #QuantumMachineLearning #QuantumAI #LifeSciences #DeepTech

To what extent can one take #quantummachinelearning models and replace them by classical surrogates when evaluating? This is explored here.

In our #quantummachinelearning seminar, @IMathYou2 gives a beautiful seminar and leads a good discussion on “some types of advantage from quantum data and its processing”.

In our #quantummachinelearning reading group, @junyuliu4 gives a wonderfully inspiring talk on representation learning via quantum neural tangent kernels.

A talk on what could be called a rigorous approach to #quantummachinelearning at the 751. WE-#Heraeus-Seminar.

In our #quantummachinelearning reading group, @AmiraMorphism gives an excellent talk on quantum backpropagation, information reuse and cheating measurement collapse.

Companies Win 1 Million Euro Contract to Use Quantum Machine Learning to Improve Climate Models 👇 Read here: #quantummachinelearning #climatemodels #quantumtechnology #innovation

Today we present the first demonstration of a provable exponential advantage in #QuantumMachineLearning over classical algorithms that is robust even on today's noisy hardware. Learn all about it at

Motivated by the role such quantum circuits play in notions of variational #quantummachinelearning, we ask: Can one learn the output distribution of a #Clifford #quantumcircuit? Yes. Can one do so for a single added #TGate? No.

This is a piece of work I am particularly happy with: It proves generalization bounds for parametrized quantum circuits in #quantummachinelearning taking data-encoding seriously. Thanks to @rndm_wlks, @jj_xyz, @EliesMiquel and particularly to @IMathYou2.

Having a great day discussing #quantummachinelearning with @WojciechSamek, @EliesMiquel and Thomas Wiegand at my second affiliation, the @FraunhoferHHI.