Quantum Machine Learning conference 2021

The first "Quantum Machine Learning" conference was organized jointly by Quantum AI Foundation, QPoland, QIndia and CEBT-SBM(MBA) - Christ University Pune Lavasa.
The event took place online on 23.10.2021.

Video recording: https://www.youtube.com/watch?v=cAsnyty3HnU 


Agenda (times in UTC+2):




Abstracts:


10:30: Introduction to the workshop (Paweł Gora, Anshul Saxena, Akash Kundu, Rishi Sreedhar).


Introduction to the workshop, presentation of lectures and speakers, presentation of the Quantum AI Foundation, QWorld, QPoland, QIndia, Christ University and CEBT.



11:00: Towards Quantum Intelligence (Amlan  Chakrabarti).


The superiority of quantum computers lies in information processing power over classical counterparts. Quantum computers’ use case could involve immediately defeating standard encryption technologies rendering them obsolete, comparing new pharmaceuticals for testing within seconds giving a significant speed-to-market advantage and increasing the speed of trading and transactions putting challenges to regulators to effectively monitor security activities. There are certain computationally hard problems in computer science, for which there exist only exponential-time classical algorithms. However, it has been proved that a class of such problems can have polynomial speedup in quantum machines due to the inherent principles of atomic computation. The intrinsic features of quantum machines like superposition, interference and entanglement promises to initiate a new platform of computing paradigm, which can be coined as “Quantum Intelligence”. 


12:00: Quantum Machine Learning - using Pennylane (Manish Gupta).


With the recent success of artificial neural networks and the emergence of Near Intermediate Scale Quantum computers a new field of quantum machine learning was established. There is hope that quantum computers will be able to build better and faster machine learning models. One of the quantum machine learning approaches is to embed quantum computers into neural network processing graphs. During the talk, I will present the basic concepts of quantum neural networks, their training, and their application in a multi-class supervised classification setting. I will show how to use Pennylane to perform training and inference using a particular type of quantum neural network. 


13:00: Variational quantum algorithms (Oskar Słowik).


Variational quantum algorithms (VQAs) are a class of hybrid quantum/classical algorithms proposed to solve some optimization problems on Noisy Intermediate-Scale Quantum devices. During my talk, I will briefly explain how VQAs work, show some examples of them (such as Variational Quantum Eigensolver and Quantum Approximate Optimization Algorithm) and discuss their possible applications and challenges related to them.


14:00: Qiskit QML (Kavitha Yogaraj).


In this tutorial let's talk the walk with coding using Qiskit QML Modules.

- Quantum Kernel module for classification & Clustering: In quantum kernel machine learning, let's learn how to map data to higher dimensional space & read the kernel matrix with train & test data.

- QML for Classification using Quantum Neural Networks (QNN) & Torch Connector: Let's learn how to use Parameterised Quantum Circuits & Torch Connector class from Qiskit Machine Learning into a PyTorch workflow 


16:00: Using PennyLane for Quantum Differentiable Programming (Antal Szava).

Advances in deep learning have greatly affected quantum computation, allowing a plethora of previously intractable scientific problems to be within reach for investigation via scientific computing. A key factor behind this new field is the availability of quantum software and differentiable programming. PennyLane is a cross-platform Python library for differentiable programming of quantum computers. It allows coding up solutions to problems in quantum machine learning, quantum chemistry and quantum computation and allows training quantum computers the same way as classical neural networks. To achieve all this, PennyLane integrates with multiple machine learning libraries and is device-agnostic: it only takes a single line of code to be changed to run the same program on simulators or real quantum hardware. During this introduction and tutorial session, you'll start with creating basic building blocks of differentiable quantum nodes, look at solutions to specific problems in the domains of quantum machine learning and quantum computation and even implement your custom quantum device in PennyLane!


Registration form (will be open until 22.10.2021, EoD CET): 

https://docs.google.com/forms/d/e/1FAIpQLSc1moTRLF7LlOkPrpl8z5AE3QLTQgneGTafU_g9k4S6duqBjQ/viewform 


Organizers: