ChemicalQDeviceMIT and Google QML Platforms as Candidates for Improving Medical
Dear Healthcare Colleague, my name is Kevin Kawchak, and I am the Founder CEO of ChemicalQDevice. Today is April 20th, 2023 for "MIT and Google QML Platforms as candidates for Improving Medical Image Analysis"
Quantum based platforms such as MIT's Quantum ML System and Google's Tensorflow Quantum support the development of image based applications.
The 4 Areas of MIT's Quantum ML System utilize Quantum Computer Systems and Machine Learning. The team's goal is to Quote "Leverage algorithm-system co-design methodology, especially with the help of Machine Learning to improve noise-robust, efficiency, and accuracy of quantum circuits on real quantum devices." [1]
MIT Area #1: TorchQuantum is an open-source library for easy construction of parameterized quantum circuits such as Quantum Neural Networks in PyTorch. This also includes batch mod interface and training on GPUs and CPUs. [2]
Area #2: QuantumNAS, short for Quantum Noise-Adaptive Search, is a comprehensive framework for noise-adaptive co-search of the variational circuit and qubit mapping. For QML, QuantumNAS was the first to achieve over 95% binary, 85% 4-class, and 32% 10-class classification accuracies on a real quantum computer. [3]
Area #3: QuantumNAT, or Quantum Noise-Aware Training, is a Post Quantum Cryptography framework to perform noise-aware optimizations in both training and inference stages for improving robustness. QuantumNAT achieved over 94% binary, 80% 4-class, and 34% 10-class classification accuracies on real quantum hardware. [4]
Area #4: QOC, or Quantum On-Chip Training, is the first experimental demonstration of practical training with parameter shift of this type. For Quantum Neural Networks, the on-chip training achieved over 90% and 60% accuracy for Binary and 4-class image classification tasks on real quantum computers. [5]
Next Platform: Google's TensorFlow Quantum, or TFQ for short, is a quantum machine learning library for rapid prototyping of hybrid quantum-classical ML models, with a focus on quantum data. The platform integrates quantum computing algorithms and logic designed in Google Cirq. In addition, existing TensorFlow APIs are provided quantum computing primitives. [6]
The 2021 article titled "TensorFlow Quantum: A Software Framework for Quantum Machine Learning" by Google Quantum AI Researchers and 24 additional organizations provides an overview of the platform. A Quote from the authors regarding TFQ includes: "This framework offers high-level abstractions for the design and training of both discriminative and generative quantum models under TensorFlow and supports high-performance quantum circuit simulators." [7]
TFQ functionalities via several basic applications include supervised learning for quantum classification, quantum control, simulating noisy quantum circuits, and quantum approximate optimization. In addition, TFQ can be used for advanced tasks including layerwise learning, Hamiltonian learning, and variational quantum eigensolvers." [7]
In summary, the MIT Quantum ML System's platform open source library, adaptive search, and two training method developments are supported by literature and appear actively engaged by researchers for prospective developments. Google's TensorFlow Quantum builds on the popular TensorFlow Machine Learning platform, and supports a wide range of quantum algorithm endeavors likely suitable for emerging radiology and/or medical image applications.
References are available in the section below. Please leave a comment regarding medical images or quantum computing. Have a productive rest of your day. Smile.
4/20/23 "MIT and Google QML Platforms as candidates for Improving Medical Image Analysis"
References: [1] https://qmlsys.mit.edu/[2] https://github.com/mit-han-lab/torchquantum[3] https://arxiv.org/abs/2107.10845[4] https://arxiv.org/abs/2110.11331[5] https://arxiv.org/abs/2202.13239[6] https://www.tensorflow.org/quantum[7] https://arxiv.org/abs/2003.02989
Created by Kevin Kawchak Founder CEO ChemicalQDevice2023 San Diego, CaliforniaAdvancing Neuroimaging
Quantum based platforms such as MIT's Quantum ML System and Google's Tensorflow Quantum support the development of image based applications.
The 4 Areas of MIT's Quantum ML System utilize Quantum Computer Systems and Machine Learning. The team's goal is to Quote "Leverage algorithm-system co-design methodology, especially with the help of Machine Learning to improve noise-robust, efficiency, and accuracy of quantum circuits on real quantum devices." [1]
MIT Area #1: TorchQuantum is an open-source library for easy construction of parameterized quantum circuits such as Quantum Neural Networks in PyTorch. This also includes batch mod interface and training on GPUs and CPUs. [2]
Area #2: QuantumNAS, short for Quantum Noise-Adaptive Search, is a comprehensive framework for noise-adaptive co-search of the variational circuit and qubit mapping. For QML, QuantumNAS was the first to achieve over 95% binary, 85% 4-class, and 32% 10-class classification accuracies on a real quantum computer. [3]
Area #3: QuantumNAT, or Quantum Noise-Aware Training, is a Post Quantum Cryptography framework to perform noise-aware optimizations in both training and inference stages for improving robustness. QuantumNAT achieved over 94% binary, 80% 4-class, and 34% 10-class classification accuracies on real quantum hardware. [4]
Area #4: QOC, or Quantum On-Chip Training, is the first experimental demonstration of practical training with parameter shift of this type. For Quantum Neural Networks, the on-chip training achieved over 90% and 60% accuracy for Binary and 4-class image classification tasks on real quantum computers. [5]
Next Platform: Google's TensorFlow Quantum, or TFQ for short, is a quantum machine learning library for rapid prototyping of hybrid quantum-classical ML models, with a focus on quantum data. The platform integrates quantum computing algorithms and logic designed in Google Cirq. In addition, existing TensorFlow APIs are provided quantum computing primitives. [6]
The 2021 article titled "TensorFlow Quantum: A Software Framework for Quantum Machine Learning" by Google Quantum AI Researchers and 24 additional organizations provides an overview of the platform. A Quote from the authors regarding TFQ includes: "This framework offers high-level abstractions for the design and training of both discriminative and generative quantum models under TensorFlow and supports high-performance quantum circuit simulators." [7]
TFQ functionalities via several basic applications include supervised learning for quantum classification, quantum control, simulating noisy quantum circuits, and quantum approximate optimization. In addition, TFQ can be used for advanced tasks including layerwise learning, Hamiltonian learning, and variational quantum eigensolvers." [7]
In summary, the MIT Quantum ML System's platform open source library, adaptive search, and two training method developments are supported by literature and appear actively engaged by researchers for prospective developments. Google's TensorFlow Quantum builds on the popular TensorFlow Machine Learning platform, and supports a wide range of quantum algorithm endeavors likely suitable for emerging radiology and/or medical image applications.
References are available in the section below. Please leave a comment regarding medical images or quantum computing. Have a productive rest of your day. Smile.
4/20/23 "MIT and Google QML Platforms as candidates for Improving Medical Image Analysis"
References: [1] https://qmlsys.mit.edu/[2] https://github.com/mit-han-lab/torchquantum[3] https://arxiv.org/abs/2107.10845[4] https://arxiv.org/abs/2110.11331[5] https://arxiv.org/abs/2202.13239[6] https://www.tensorflow.org/quantum[7] https://arxiv.org/abs/2003.02989
Created by Kevin Kawchak Founder CEO ChemicalQDevice2023 San Diego, CaliforniaAdvancing Neuroimaging