Hybrid Quantum–Classical Neural Networks for Efficient MNIST Binary Image Classification
Abstract
:1. Introduction
1.1. Motivation and Contribution
- We presents a novel Hybrid Quantum–Classical Neural Network (H-QNN) model. This architecture integrates fundamentals feature mapping with a classical neural network to effectively improve image classification tasks, with a specific focus on binary classification using the MNIST dataset.
- The proposed work refines the quantum layer by using parameterized quantum circuits (VQCs) that contain RY rotation gates and CX entanglement gates, in conjunction with the ZZFeature Map for efficient data encoding. This design reduces the circuit depth while maintaining favorable computational efficiency.
- By obtaining a score of 99.7% on the binary MNIST classification task, the proposed H-QNN model shows that it is more accurate and needs a lot less computing power than traditional CNNs and QCNNs.
- This paper underscores the broader applicability of hybrid quantum–classical models in different domains like finance, cybersecurity, and medical diagnostics and highlights the potential for scaling these models to handle more complex data and tasks.
1.2. Organization
2. Related Work
2.1. Convolutional Neural Networks (CNNs)
2.2. Quantum Neural Networks (QNNs)
2.3. Hybrid Quantum–Classical Neural Networks (HQNNs)
3. Basic Preliminaries
3.1. Superposition
3.2. Qubits
3.3. Quantum Gates
3.4. Entanglement
3.5. Measurement
3.6. CNNs and QNNs: Foundations, Architectures, and Recent Advancements in Quantum Machine Learning
4. Materials and Methodology for Proposed Method
4.1. MNIST Dataset Description
4.2. Hybrid Quantum-Classical Architecture
- Quantum Layer: The quantum layer serves as a bridge between classical and quantum processing by transforming classical data into quantum states. Using quantum feature maps, it encodes data into a higher-dimensional quantum space, enabling complex relationships to be captured that might be hidden in classical representation. This mapping enhances the model’s ability to identify intricate patterns by leveraging quantum properties, ultimately improving the performance of hybrid quantum–classical algorithms. Through this transformation, classical inputs gain access to the computational advantages of quantum mechanics.Quantum Feature Mapping: We use the ZZFeature map to encode the conventional feature vector into a quantum state. Quantum states process information through unitary gates by enabling the model to learn more complex and non-linear patterns than classical models. After encoding, the quantum state is passed to the variational quantum circuit (VQC) for further processing.
- Variational Quantum Circuit (VQC): A VQC is a quantum circuit with parameterized quantum gates, forming the foundation of the quantum layer in hybrid quantum–classical models. These gates contain trainable variables, allowing the circuit’s parameters to be adjusted. During training, these parameters are iteratively optimized based on a cost function, enabling the VQC to learn patterns or solutions. This iterative process allows VQCs to approximate complex functions and contribute to quantum machine learning and optimization tasks.
- Parametric Quantum Gates: The VQC employs RY rotation gates to rotate qubit states on the Y-axis of the Bloch sphere and CX entanglement gates to establish qubit correlations.
- Real Amplitudes Ansatz: This ansatz is employed in our VQC design, which includes RY rotations and CX gates in sequential layers. The circuit depth and number of trainable parameters can be adjusted based on the complexity of the problem. For efficiency and to capture complex patterns, only one repetition of the circuit is used. During training, classical optimization methods such as the Adam optimizer are applied to minimize the loss function and align the VQC’s output with the target labels.
- Classical Layer: The VQC produces quantum measurements, which are then processed by a classical layer to make decisions:
- Measurement: Quantum circuits are measured to collapse quantum states into classical data, which yield the system’s state probabilities. These probabilities are then fed into the classical model for further processing.
- Classical Neural Network: A feed-forward neural network (or another classifier like SVM) takes the quantum probabilities as input, and produces a binary output (‘0’ or ‘1’). It allows the model to combine quantum feature encoding with classical decision-making.
4.3. Dataset Preprocessing
4.3.1. Data Splitting
4.3.2. Standard Scaling
4.4. Quantum Circuit Design
4.4.1. State Preparation (ZZFeature Map)
- To encode data, the ZZFeature Map was used, inspired by traditional machine learning kernel methods. This approach non-linearly transforms classical datasets into a higher-dimensional quantum space.
- The feature map generates quantum states. The function states describe a larger feature space that enables the classifier to spot a separating hyperplane in the extended quantum area. The unitary operation creates a circuit of circuit of Hadamard gates (H) interleaved with entangling gates to achieve the encoding.
- This approach improves the learning of non-linear data patterns through the use of quantum feature mapping.
4.4.2. Real Amplitudes Ansatz
- This ansatz is made of single-qubit rotation gates and two-qubit entanglement gates . First, the parameterized gates are applied to each qubit, and then the entangling gates are executed.
- A second parameterized rotation is performed as a fourth step after the entanglement wall. This setup allows for slow and accurate encoding of data into the quantum states. Moreover, circuit depth was reduced by selecting a single repetition of the gates and rotations.
Algorithm 1 Hybrid quantum–classical module. |
|
4.5. Optimization
5. Experimental Setup
- MNIST Dataset: It starts with MNIST, which consists of handwritten digit images.Subsampling Classes 0 and 1: A portion of the MNIST data is chosen, composed solely of images of the digits “0” and “1”. This simplifies the classification task to a binary classification. Labeled and predicted images from the MNIST dataset are shown in Figure 7.
- Data Splitting: The subsampled dataset is divided into two sets: training set (i.e., this includes all records used to train the model) and testing set (i.e., a portion is used for validations and other part is reserved for model validation after construction and development).
- Simulation (Training and Testing): The training set introduces the hybrid quantum classical model. It refines weights and quantum circuit parameters to improve image classification. After that, the test set evaluates the model’s generalization by running it on new data to assess its accuracy in predicting images as “0” or “1”.
- Libraries used: The following libraries were used for the simulation:PyTorch: A classical framework widely used for training neural networks data processing tasksScikit-Learn (Sk-Learn): A Python 13.11.0 library for dataset splitting, model evaluation, and classical machine learning.Qiskit: It is an IBM programming framework utilized to build and simulat quantum circuits in a hybrid quantum–classical architecture.
- Running on Jupyter Notebook: All computation, data processing, training, testing, and evaluation were conducted in Jupyter Notebook, which is widely used for running Python code interactively.
- Evaluation Metrics: After testing the model, several evaluation metrics were computed to assess the model’s performance:Training Time: measures the duration taken to complete the model training.Accuracy: The proportion of correctly identified images by classifier out of total testing set.F1-Score: The average mean of precision and recall that is useful for handling class imbalance.Precision: The percentage of true positive identifications, e.g., correctly identifying images of “1”.Recall: The ratio of true positive identifications to all actual positive cases in the testing dataset.
6. Results and Discussion
6.1. Performance Metrics
6.2. Accuracy Metrics
6.3. Confusion Matrix
6.4. Comparison with CNN and QCNN
- CNNs have been reported to give very good results when used with MNIST dataset with an accuracy of 99.2% with LeNet architecture [54]. Although CNNs easily learn stacked features using convolutional layers, they may be brittle to other forms of features compared to these more complex networks. However, to overcome this limitation H-QNN uses the concept of the quantum feature mapping so that it can learn the complex representations effectively.
- ResNet, which uses residual connection, gives a 99.6% accuracy of the MNIST. It performs well in training deep networks due to helping solve the vanishing gradient problem but is precluded by classical computation that limits its ability to learn intricate data representation patterns [53]. Like other quantum neural networks, H-QNN takes advantage of parallelism within the quantum computer, thus making it capable of solving more complex patterns in a shorter time than is possible in classical computers.
- DenseNet achieves an accuracy of 99.7% by dense connections enabled with efficient feature reuse and gradient flow. While this architecture is very powerful in many ways, it requires more memory and more computation than the network depth grows [53,54]. By allowing quantum circuits to perform more effective complex feature mapping, H-QNN obtains similar accuracy with fewer resources.Furthermore, self-attention mechanisms are used in Vision Transformers (ViT) and are able to work better with larger datasets. Despite such accuracy ranges between 98.6% and 99.2% on MNIST. ViT is computationally expensive and lacks specialization to smaller datasets [55,56]. Compared to datasets of medium size, H-QNN is particularly efficient at extracting rich information using quantum features.
- CapNet preserves spatial hierarchies and has strong performance, which achieves 99.65% accuracy on MNIST. Despite that, CapNet’s dynamic routing is a hindrance to training and adds computational load [57]. Its hybrid approach, similar to H-QNN, achieves the same accuracy in a lower complexity.
6.5. Discussion
7. Conclusions and Future Work
Author Contributions
Funding
Data Availability Statement
Conflicts of Interest
References
- Lu, D.; Weng, Q. A survey of image classification methods and techniques for improving classification performance. Int. J. Remote Sens. 2007, 28, 823–870. [Google Scholar] [CrossRef]
- Wang, S.Y.; Liao, W.S.; Hsieh, L.C.; Chen, Y.Y.; Hsu, W.H. Learning by expansion: Exploiting social media for image classification with few training examples. Neurocomputing 2012, 95, 117–125. [Google Scholar] [CrossRef]
- Turay, T.; Vladimirova, T. Toward performing image classification and object detection with convolutional neural networks in autonomous driving systems: A survey. IEEE Access 2022, 10, 14076–14119. [Google Scholar] [CrossRef]
- Schuld, M.; Sinayskiy, I.; Petruccione, F. An introduction to quantum machine learning. Contemp. Phys. 2015, 56, 172–185. [Google Scholar] [CrossRef]
- Ciliberto, C.; Herbster, M.; Ialongo, A.D.; Pontil, M.; Rocchetto, A.; Severini, S.; Wossnig, L. Quantum machine learning: A classical perspective. Proc. R. Soc. A Math. Phys. Eng. Sci. 2018, 474, 20170551. [Google Scholar] [CrossRef]
- Alpaydin, E. Machine Learning; MIT Press: Cambridge, MA, USA, 2021. [Google Scholar]
- Jordan, M.I.; Mitchell, T.M. Machine learning: Trends, perspectives, and prospects. Science 2015, 349, 255–260. [Google Scholar] [CrossRef]
- Carleo, G.; Cirac, I.; Cranmer, K.; Daudet, L.; Schuld, M.; Tishby, N.; Vogt-Maranto, L.; Zdeborová, L. Machine learning and the physical sciences. Rev. Mod. Phys. 2019, 91, 045002. [Google Scholar] [CrossRef]
- DiVincenzo, D.P. Quantum computation. Science 1995, 270, 255–261. [Google Scholar] [CrossRef]
- Vedral, V.; Plenio, M.B. Basics of quantum computation. Prog. Quantum Electron. 1998, 22, 1–39. [Google Scholar] [CrossRef]
- Nielsen, M.A.; Chuang, I.L. Quantum Computation and Quantum Information; Cambridge University Press: Cambridge, UK, 2010. [Google Scholar]
- Sim, S.; Johnson, P.D.; Aspuru-Guzik, A. Expressibility and entangling capability of parameterized quantum circuits for hybrid quantum-classical algorithms. Adv. Quantum Technol. 2019, 2, 1900070. [Google Scholar] [CrossRef]
- Leyton-Ortega, V.; Perdomo-Ortiz, A.; Perdomo, O. Robust implementation of generative modeling with parametrized quantum circuits. Quantum Mach. Intell. 2021, 3, 17. [Google Scholar] [CrossRef]
- Gupta, S.; Zia, R. Quantum neural networks. J. Comput. Syst. Sci. 2001, 63, 355–383. [Google Scholar] [CrossRef]
- Kiss, O.; Tacchino, F.; Vallecorsa, S.; Tavernelli, I. Quantum neural networks force fields generation. Mach. Learn. Sci. Technol. 2022, 3, 035004. [Google Scholar] [CrossRef]
- Hafeez, M.A.; Munir, A.; Ullah, H. H-QNN: A Hybrid Quantum–Classical Neural Network for Improved Binary Image Classification. AI 2024, 5, 1462–1481. [Google Scholar] [CrossRef]
- Cheng, D.; Chen, L.; Lv, C.; Guo, L.; Kou, Q. Light-guided and cross-fusion U-Net for anti-illumination image super-resolution. IEEE Trans. Circuits Syst. Video Technol. 2022, 32, 8436–8449. [Google Scholar] [CrossRef]
- Yu, S.; Guan, D.; Gu, Z.; Guo, J.; Liu, Z.; Liu, Y. Radar Target Complex High-Resolution Range Profile Modulation by External Time Coding Metasurface. IEEE Trans. Microw. Theory Tech. 2024, 72, 6083–6093. [Google Scholar] [CrossRef]
- Wang, G.; Li, R. DSolving: A novel and efficient intelligent algorithm for large-scale sliding puzzles. J. Exp. Theor. Artif. Intell. 2017, 29, 809–822. [Google Scholar] [CrossRef]
- Guo, T.; Yuan, H.; Hamzaoui, R.; Wang, X.; Wang, L. Dependence-Based Coarse-to-Fine Approach for Reducing Distortion Accumulation in G-PCC Attribute Compression. IEEE Trans. Ind. Inform. 2024, 20, 11393–11403. [Google Scholar] [CrossRef]
- Xu, Y.; Ding, L.; He, P.; Lu, Z.; Zhang, J. A Memory-Efficient Tri-Stage Polynomial Multiplication Accelerator Using 2D Coupled-BFUs. IEEE Trans. Circuits Syst. Regul. Pap. 2024. [Google Scholar] [CrossRef]
- Mi, B.; Huang, D.; Wan, S.; Hu, Y.; Choo, K.K.R. A post-quantum light weight 1-out-n oblivious transfer protocol. Comput. Electr. Eng. 2019, 75, 90–100. [Google Scholar] [CrossRef]
- Li, W.; Chu, P.C.; Liu, G.Z.; Tian, Y.B.; Qiu, T.H.; Wang, S.M. An image classification algorithm based on hybrid quantum classical convolutional neural network. Quantum Eng. 2022, 2022, 5701479. [Google Scholar] [CrossRef]
- Fan, F.; Shi, Y.; Guggemos, T.; Zhu, X.X. Hybrid quantum-classical convolutional neural network model for image classification. IEEE Trans. Neural Netw. Learn. Syst. 2023. [Google Scholar] [CrossRef] [PubMed]
- LeCun, Y.; Boser, B.; Denker, J.S.; Henderson, D.; Howard, R.E.; Hubbard, W.; Jackel, L.D. Backpropagation applied to handwritten zip code recognition. Neural Comput. 1989, 1, 541–551. [Google Scholar] [CrossRef]
- Zhang, C.W.; Yang, M.Y.; Zeng, H.J.; Wen, J.P. Pedestrian detection based on improved LeNet-5 convolutional neural network. J. Algorithms Comput. Technol. 2019, 13, 1748302619873601. [Google Scholar] [CrossRef]
- Yuan, Z.W.; Zhang, J. Feature extraction and image retrieval based on AlexNet. In Proceedings of the Eighth International Conference on Digital Image Processing (ICDIP 2016), Chengu, China, 20–22 May 2016; SPIE: Bellingham, WA, USA, 2016; Volume 10033, pp. 65–69. [Google Scholar]
- Sengupta, A.; Ye, Y.; Wang, R.; Liu, C.; Roy, K. Going deeper in spiking neural networks: VGG and residual architectures. Front. Neurosci. 2019, 13, 95. [Google Scholar] [CrossRef]
- Al-Qizwini, M.; Barjasteh, I.; Al-Qassab, H.; Radha, H. Deep learning algorithm for autonomous driving using googlenet. In Proceedings of the 2017 IEEE Intelligent Vehicles Symposium (IV), Los Angeles, CA, USA, 11–14 June 2017; IEEE: Piscataway, NJ, USA, 2017; pp. 89–96. [Google Scholar]
- Zhu, C. Computational intelligence-based classification system for the diagnosis of memory impairment in psychoactive substance users. J. Cloud Comput. 2024, 13, 119. [Google Scholar] [CrossRef]
- Wu, Z.; Shen, C.; Van Den Hengel, A. Wider or deeper: Revisiting the resnet model for visual recognition. Pattern Recognit. 2019, 90, 119–133. [Google Scholar] [CrossRef]
- Song, W.; Wang, X.; Jiang, Y.; Li, S.; Hao, A.; Hou, X.; Qin, H. Expressive 3D Facial Animation Generation Based on Local-to-global Latent Diffusion. IEEE Trans. Vis. Comput. Graph. 2024, 30, 7397–7407. [Google Scholar] [CrossRef] [PubMed]
- Levi, G.; Hassner, T. Age and gender classification using convolutional neural networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, Boston, MA, USA, 7–12 June 2015; pp. 34–42. [Google Scholar]
- Karpathy, A.; Toderici, G.; Shetty, S.; Leung, T.; Sukthankar, R.; Fei-Fei, L. Large-scale video classification with convolutional neural networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Columbus, OH, USA, 23–28 June 2014; pp. 1725–1732. [Google Scholar]
- Jmour, N.; Zayen, S.; Abdelkrim, A. Convolutional neural networks for image classification. In Proceedings of the 2018 International Conference on Advanced Systems and Electric Technologies (IC_ASET), Hammamet, Tunisia, 22–25 March 2018; IEEE: Piscataway, NJ, USA, 2018; pp. 397–402. [Google Scholar]
- Kang, L.; Kumar, J.; Ye, P.; Li, Y.; Doermann, D. Convolutional neural networks for document image classification. In Proceedings of the 2014 22nd International Conference on Pattern Recognition, Stockholm, Sweden, 24–28 August 2014; IEEE: Piscataway, NJ, USA, 2014; pp. 3168–3172. [Google Scholar]
- Sermanet, P.; Chintala, S.; LeCun, Y. Convolutional neural networks applied to house numbers digit classification. In Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012), Tsukuba, Japan, 11–15 November 2012; IEEE: Piscataway, NJ, USA, 2012; pp. 3288–3291. [Google Scholar]
- Wu, H. CNN-Based Recognition of Handwritten Digits in MNIST Database. In Research School of Computer Science; The Australia National University: Canberra, Australia, 2018. [Google Scholar]
- Palvanov, A.; Im Cho, Y. Comparisons of deep learning algorithms for MNIST in real-time environment. Int. J. Fuzzy Log. Intell. Syst. 2018, 18, 126–134. [Google Scholar] [CrossRef]
- Jeswal, S.; Chakraverty, S. Recent developments and applications in quantum neural network: A review. Arch. Comput. Methods Eng. 2019, 26, 793–807. [Google Scholar] [CrossRef]
- Nguyen, T.; Paik, I.; Watanobe, Y.; Thang, T.C. An evaluation of hardware-efficient quantum neural networks for image data classification. Electronics 2022, 11, 437. [Google Scholar] [CrossRef]
- Zhou, R.; Ding, Q. Quantum mp neural network. Int. J. Theor. Phys. 2007, 46, 3209–3215. [Google Scholar] [CrossRef]
- Safari, A.; Ghavifekr, A.A. Quantum neural networks (QNN) application in weather prediction of smart grids. In Proceedings of the 2021 11th Smart Grid Conference (SGC), Tabriz, Iran, 7–9 December 2021; IEEE: Piscataway, NJ, USA, 2021; pp. 1–6. [Google Scholar]
- Paquet, E.; Soleymani, F. QuantumLeap: Hybrid quantum neural network for financial predictions. Expert Syst. Appl. 2022, 195, 116583. [Google Scholar] [CrossRef]
- Wei, S.; Chen, Y.; Zhou, Z.; Long, G. A quantum convolutional neural network on NISQ devices. AAPPS Bull. 2022, 32, 2. [Google Scholar] [CrossRef]
- Kadam, S.S.; Adamuthe, A.C.; Patil, A.B. CNN model for image classification on MNIST and fashion-MNIST dataset. J. Sci. Res. 2020, 64, 374–384. [Google Scholar] [CrossRef]
- Xia, R.; Kais, S. Hybrid quantum-classical neural network for calculating ground state energies of molecules. Entropy 2020, 22, 828. [Google Scholar] [CrossRef]
- Hellstem, G. Hybrid quantum network for classification of finance and MNIST data. In Proceedings of the 2021 IEEE 18th International Conference on Software Architecture Companion (ICSA-C), Stuttgart, Germany, 22–26 March 2021; IEEE: Piscataway, NJ, USA, 2021; pp. 1–4. [Google Scholar]
- Bokhan, D.; Mastiukova, A.S.; Boev, A.S.; Trubnikov, D.N.; Fedorov, A.K. Multiclass classification using quantum convolutional neural networks with hybrid quantum-classical learning. Front. Phys. 2022, 10, 1069985. [Google Scholar] [CrossRef]
- Xu, Z.; Hu, Y.; Yang, T.; Cai, P.; Shen, K.; Lv, B.; Chen, S.; Wang, J.; Zhu, Y.; Wu, Z.; et al. Parallel Structure of Hybrid Quantum-Classical Neural Networks for Image Classification. Res. Sq. 2024; preprint. [Google Scholar]
- Ling, Y.Q.; Zhang, J.H.; Zhang, L.H.; Li, Y.R.; Huang, H.L. Image Classification Using Hybrid Classical-Quantum Neutral Networks. Int. J. Theor. Phys. 2024, 63, 125. [Google Scholar] [CrossRef]
- Islam, M.; Chowdhury, M.; Khan, Z.; Khan, S.M. Hybrid quantum-classical neural network for cloud-supported in-vehicle cyberattack detection. IEEE Sensors Lett. 2022, 6, 1–4. [Google Scholar] [CrossRef]
- Sarmah, J.; Saini, M.L.; Kumar, A.; Chasta, V. Performance Analysis of Deep CNN, YOLO, and LeNet for Handwritten Digit Classification. In Proceedings of the International Conference on Artificial Intelligence on Textile and Apparel, Bangalore, India, 11–12 August 2023; Springer: Singapore, 2023; pp. 215–227. [Google Scholar]
- Choudhuri, A.R.; Thakurata, B.G.; Debnath, B.; Ghosh, D.; Maity, H.; Chattopadhyay, N.; Chakraborty, R. MNIST Image Classification Using Convolutional Neural Networks. In Modeling, Simulation and Optimization: Proceedings of CoMSO 2021; Springer: Berlin/Heidelberg, Germany, 2022; pp. 255–266. [Google Scholar]
- Zhang, L.; Lu, J.; Zheng, S.; Zhao, X.; Zhu, X.; Fu, Y.; Xiang, T.; Feng, J.; Torr, P.H. Vision transformers: From semantic segmentation to dense prediction. Int. J. Comput. Vis. 2024, 132, 6142–6162. [Google Scholar] [CrossRef]
- Hwang, E.E.; Chen, D.; Han, Y.; Jia, L.; Shan, J. Multi-Dataset Comparison of Vision Transformers and Convolutional Neural Networks for Detecting Glaucomatous Optic Neuropathy from Fundus Photographs. Bioengineering 2023, 10, 1266. [Google Scholar] [CrossRef] [PubMed]
- Choudhary, S.; Saurav, S.; Saini, R.; Singh, S. Capsule networks for computer vision applications: A comprehensive review. Appl. Intell. 2023, 53, 21799–21826. [Google Scholar] [CrossRef]
- Zhou, G.; Li, H.; Song, R.; Wang, Q.; Xu, J.; Song, B. Orthorectification of fisheye image under equidistant projection model. Remote Sens. 2022, 14, 4175. [Google Scholar] [CrossRef]
- Prajapat, S.; Kumar, P.; Kumar, D.; Das, A.K.; Hossain, M.S.; Rodrigues, J.J. Quantum secure authentication scheme for internet of medical things using blockchain. IEEE Internet Things J. 2024, 11, 38496–38507. [Google Scholar] [CrossRef]
- Prajapat, S.; Gautam, D.; Kumar, P.; Jangirala, S.; Das, A.K.; Park, Y.; Lorenz, P. Secure lattice-based aggregate signature scheme for vehicular Ad Hoc networks. IEEE Trans. Veh. Technol. 2024, 73, 12370–12384. [Google Scholar] [CrossRef]
- Prajapat, S.; Thakur, G.; Kumar, P.; Das, A.K.; Hossain, M.S. A blockchain-assisted privacy-preserving signature scheme using quantum teleportation for metaverse environment in Web 3.0. Future Gener. Comput. Syst. 2024, 164, 107581. [Google Scholar] [CrossRef]
- Prajapat, S.; Kumar, P.; Kumar, S. A privacy preserving quantum authentication scheme for secure data sharing in wireless body area networks. Clust. Comput. 2024, 27, 9013–9029. [Google Scholar] [CrossRef]
- Zhou, G.; Wang, Q.; Huang, Y.; Tian, J.; Li, H.; Wang, Y. True2 orthoimage map generation. Remote Sens. 2022, 14, 4396. [Google Scholar] [CrossRef]
- Pan, H.; Wang, Y.; Li, Z.; Chu, X.; Teng, B.; Gao, H. A complete scheme for multi-character classification using EEG signals from speech imagery. IEEE Trans. Biomed. Eng. 2024, 71, 2454–2462. [Google Scholar] [CrossRef]
- Kumar, P.; Bharmaik, V.; Prajapat, S.; Thakur, G.; Das, A.K.; Shetty, S.; Rodrigues, J.J. A Secure and Privacy-Preserving Signature Protocol Using Quantum Teleportation in Metaverse Environment. IEEE Access 2024, 12, 96718–96728. [Google Scholar] [CrossRef]
- Xu, X.; Fu, X.; Zhao, H.; Liu, M.; Xu, A.; Ma, Y. Three-Dimensional Reconstruction and Geometric Morphology Analysis of Lunar Small Craters within the Patrol Range of the Yutu-2 Rover. Remote Sens. 2023, 15, 4251. [Google Scholar] [CrossRef]
- Prajapat, S.; Kumar, D.; Kumar, P. Quantum image encryption protocol for secure communication in healthcare networks. Clust. Comput. 2025, 28, 3. [Google Scholar] [CrossRef]
- Prajapat, S.; Dhiman, A.; Kumar, S.; Kumar, P. A practical convertible quantum signature scheme with public verifiability into universal quantum designated verifier signature using self-certified public keys. Quantum Inf. Process. 2024, 23, 331. [Google Scholar] [CrossRef]
- Yao, F.; Zhang, H.; Gong, Y. DifSG2-CCL: Image Reconstruction Based on Special Optical Properties of Water Body. IEEE Photonics Technol. Lett. 2024, 36, 1417–1420. [Google Scholar] [CrossRef]
- Shi, H.; Hayat, M.; Cai, J. Unified open-vocabulary dense visual prediction. IEEE Trans. Multimed. 2024, 26, 8704–8716. [Google Scholar] [CrossRef]
- Hu, C.; Dong, B.; Shao, H.; Zhang, J.; Wang, Y. Toward purifying defect feature for multilabel sewer defect classification. IEEE Trans. Instrum. Meas. 2023, 72, 1–11. [Google Scholar] [CrossRef]
Reference | Technique Used | Highlights | Limitations |
---|---|---|---|
Zhang et al. [26] | LeNet | In this article, an innovative LeNet-5 CNN is developed and implemented for pedestrian detection in V2X-driven systems at a 25% miss rate, surpassing SA-Fast R-CNN and the existing LeNet-5 CNN. | The shortcomings of the paper include a low miss rate of 25%, meaning that there is potential for further refinement. It is not tested on other kinds of data apart from pedestrians, and it may perform poorly on complex scenes or if a system must work faster computationally. |
Yuan and Zhang [27] | AlexNet | To improve feature extraction and assist in image retrieval, this paper employs AlexNet, which is good for capturing complicated image features, increasing retrieval precision, and surpassing more conventional image classification and retrieval techniques. | Some of the weaknesses in this work are that the large number of parameters may result in overfitting, especially with the use of AlexNet. Further, it is not efficient when the dataset size is relatively small and is not compared with more contemporary architectural designs. |
Sengupta et al. [28] | Spiking Neural Networks (SNNs) | The paper also applies and enhances SNNs to deeper architectures such as VGG and ResNet to lift the accuracy of visualization jobs like CIFAR-10 and ImageNet while utilizing lesser requests of hardware through event-driven neuromorphic design. | The main limitations of this work include implementation complexity, data dependence, generalization concern, requirement of specific hardware, and interpretability of deep spiking neural networks. |
Wu et al. [31] | Residual Network (ResNet) | Specifically, the paper evaluates network width compared to depth in certain ResNet architectures and suggests a shallow architecture over deep tasks such as image classification or segmentation. | Such weaknesses of this work consist of the absence of the generic investigation of other architectures apart from ResNet, signs of overfitting in more extensive networks, and a lack of validation across more diverse datasets. |
Jeswal and Chakraverty [40] | QNNs | The paper provides a discussion of the development of quantum neural networks (QNNs), including ways that quantum concepts are incorporated with neural structures and possible uses for QNNs in fields including image classification and search for superior learning algorithms. | The drawbacks of the work include the relatively recent development of quantum neural networks, a possible problem of scalability, a dependency on complex quantum circuits, and minimum tests on various applications. |
Nguyen et al. [41] | QNNs | This paper assesses the depth of quantum neural networks for image classification that encompasses encoding techniques and circuits with respect to the MNIST dataset of varying diverse neural networks with respect to hardware efficiency. | Some of the shortcomings of this work include issues with the ability to extend the work to bigger sets, the quantum circuit implementation problem, the lack of many experiments and instead using simulations, and an overdependence on theory and not much application of models. |
Sarmah et al. [53] | Deep CNN, YOLOv7, and LeNet-5 | This paper evaluates Deep CNN, YOLOv7, and LeNet-5 and concludes a maximum accuracy of 99.38% using LeNet-5 for handwritten digits on various datasets. | The work has some limitations, as follows: data used in this work are standard datasets, which represent simple scenarios and do not demonstrate actual real-world complexity; performance evaluation involves small sample sizes, which may lead to overfitting. |
Choudhuri et al. [54] | CNN | The paper employs convolutional neural networks, known as CNNs, when categorizing the MNIST handwritten digit, and several architectures and optimizers are examined with respect to accuracy and efficiency. | Some of the drawbacks are as follows: the used dataset is very modest, and the complexity of the real world is not reflected; overfitting may occur due to choices of the model architecture and the amount and types of the data augmentation. |
Zhang et al. [55] | ViT | The paper extends the use of vision transformers (ViTs) to challenges associated with dense prediction, such as semantic segmentation. It demonstrates that ViTs can achieve better global contextual learning than that of traditional CNNs to enhance the recognition of visuals at an enhanced level. | This work’s main weakness is the applicability of vision transformers (ViTs), which are inherently more expensive in terms of data and computational power needed for training, especially for relatively small-scale tasks, than CNNs. |
Hwang et al. [56] | ViT | The comparative study of ViT and CNN on glaucomatous optic neuropathy detection from fundus photographs of the eyes is presented in the paper. The paper shows that both the ViT and CNN are effective for medical image classification across various datasets. | The main drawback of this work is that the efficiency of vision transformers (ViTs) is high, but it requires a large amount of data and computational power. Furthermore, it can be noted that model performance is likely to be different when working with patients of different ages and sexes, as well as with images of different quality. |
Choudhary et al. [57] | CapNet | The paper briefly assesses capsule networks (CapsNets) in computer vision, highlighting their consistencies in preserving spatial hierarchies as well as the enhanced performance of CapsNets in object detection and segmentation beyond CNNs. | The weakness of this work is the consequence of the relatively large computational complexity and training time of the used capsule networks (CapsNets). They also have difficulties when working with large datasets, so it turns out that for some large-scale applications, the rate at which CNNs work is better. |
Hellstem [48] | H-QNNs | The paper offers an approach that embeds quantum circuits and classical neural networks for measuring the performance of the former on the MNIST images and financial data and how it may outperform other methods. | The drawback of this work includes the dependence on quantum computing facilities that are still under construction both in terms of technology and implementation and that it may be challenging to incorporate quantum and conventional models. |
Layer Type | Output Shape | Number of Parameters |
---|---|---|
Conv2D (1, 2, 5 × 5) | (None, 2, 24, 24) | 52 |
BatchNorm2d (2) | (None, 2, 24, 24) | 4 |
Conv2D (2, 16, 5 × 5) | (None, 16, 8, 8) | 816 |
BatchNorm2d (16) | (None, 16, 8, 8) | 32 |
Max Pooling | (None, 16, 4, 4) | 0 |
Dropout2d | (None, 16, 4, 4) | 0 |
Flatten | (None, 256) | 0 |
Dense (Linear) | (None, 64) | 16,448 |
Dense (Linear) | (None, 2) | 130 |
TorchConnector (QNN) | (None, 1) | Depends on QNN |
Dense (Linear) | (None, 1) | 2 |
Layer Type | Output Shape | Number of Parameters |
---|---|---|
ZZFeatureMap (2 qubits) | (None, 2) | 2 |
RealAmplitudes (2 qubits) | (None, 2) | 4 |
EstimatorQNN | (None, 1) | 6 |
Model | Accuracy on MNIST | Key Strengths | Key Weaknesses |
---|---|---|---|
CNN | 99.2% | Effective for hierarchical feature extraction | Struggles with complex patterns |
ResNet | 99.6% | Efficient training with residual connections | Higher computational demand |
DenseNet | 99.7% | Efficient gradient flow and feature reuse | Requires more memory |
ViT | 98.6–99.2% | Self-attention for long-range dependencies | High resource demand, requires large datasets |
CapNet | 99.65% | Preserves spatial hierarchies, handles viewpoint changes | Complex training process |
H-QNN (Our Model) | 99.7% | Quantum feature mapping, reduced computational cost | Limited by current quantum hardware |
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Ranga, D.; Prajapat, S.; Akhtar, Z.; Kumar, P.; Vasilakos, A.V. Hybrid Quantum–Classical Neural Networks for Efficient MNIST Binary Image Classification. Mathematics 2024, 12, 3684. https://doi.org/10.3390/math12233684
Ranga D, Prajapat S, Akhtar Z, Kumar P, Vasilakos AV. Hybrid Quantum–Classical Neural Networks for Efficient MNIST Binary Image Classification. Mathematics. 2024; 12(23):3684. https://doi.org/10.3390/math12233684
Chicago/Turabian StyleRanga, Deepak, Sunil Prajapat, Zahid Akhtar, Pankaj Kumar, and Athanasios V. Vasilakos. 2024. "Hybrid Quantum–Classical Neural Networks for Efficient MNIST Binary Image Classification" Mathematics 12, no. 23: 3684. https://doi.org/10.3390/math12233684
APA StyleRanga, D., Prajapat, S., Akhtar, Z., Kumar, P., & Vasilakos, A. V. (2024). Hybrid Quantum–Classical Neural Networks for Efficient MNIST Binary Image Classification. Mathematics, 12(23), 3684. https://doi.org/10.3390/math12233684