Quantum Machine Learning

Welcome readers, this is my 4th blog in the Quantum Computing series, in this blog I will be discussing the application of quantum computing in machine learning.

Quantum Machine learning can be categorized based on two types data generating system and data processing device (See Figure 1).

1. CC algorithms/Quantum-inspired machine learning/Dequantizing algorithms
2. QC algorithms/Machine learning for quantum algorithms
3. CQ algorithms/quantum for machine learning
4. QQ algorithms/quantum generalized machine learning

Figure 1

CC algorithms/Quantum-inspired machine learning

In quantum-inspired machine learning, the generating device is classical, and the data processing device is also classical, but the difference is in solving certain problems in machine learning, algorithms are developed based on the inspiration from Quantum mechanics.

For example: - In 2016, Kerenidis and Prakash proposed a quantum recommendation system. They proposed a quantum algorithm for a recommendation system that has running time O(poly(k)polylog(mn)).

If you are more interested in reading about their work, you can check this link.

Classical Algorithms which are inspired by quantum principles:

1. Quantum Inspired machine learning
2. Dequantizing algorithm

Here Dequantizing algorithm means researchers are defining a classical model that assumes the same level of strength input using quantum input assumption and achieving the accuracy with what quantum computers have achieved.

QC algorithms/machine learning for quantum algorithms

Use machine learning to improve quantum systems, In this data is generated by quantum systems and using machine learning we can improve quantum systems

Examples:

1. Classical machine learning can help in correcting quantum errors generated in quantum circuits.
2. We can use neural networks for mimicking quantum systems.

CQ Algorithms/Quantum for machine learning

In this category, we use quantum computers to improve machine learning algorithms or solve the problems of machine learning through quantum computing.

In this category, data is generated by classical systems and that data is processed on quantum computers.

With this machine learning problems can be solved efficiently faster.

QQ Algorithms/quantum generalized machine learning

In this category of algorithms, data generated is from quantum systems and data is processed also in quantum systems.

Examples of input data:

1). Large Hadron Collider (LHC)

2). Quantum Computer

In this set of categories, we are using quantum computers for solving tasks of machine learning.

The best example of Quantum Generalized machine learning on the data generated produced by the Large Hadron Collider is the effectively classify b-jets, jet originating from b-quarks from proton-proton collisions in the LHCb experiment.

For more details about this approach, you can refer to this link.

That all, this is my short blog on how quantum computing can be used as an application in machine learning. If you want to read my previous blogs on Quantum Computing Series, Please follow the below links. And if you liked the blogs, do share and clap the blogs.

Introduction to Quantum Computing: Classical Mechanics vs Quantum Mechanics

Classical Bit vs Quantum Qubit

Creating, Retaining and reading out Qubit

https://www.linkedin.com/in/jugal-oza-974048109/