Adaptive Deep Neural Network Optimization : Algorithmic and Architectural Frameworks = 알고리즘 및 아키텍처 프레임워크 분석을 통한 적응형 심층 신경망 최적화

다국어 초록 (Multilingual Abstract)

Spurred by recent advancements in machine learning, neural networks (NNs) are widely utilized in academia and industries. Especially, these NNs show great performance in analyzing complex datasets and reducing the losses in given tasks with various optimization techniques. This rapid advancement is boosted by increasing storage and computation capabilities. These capabilities enable the NNs to use increasingly larger and heavier structures to achieve high performance. Consequently, recent studies and industrial applications aim to increase the size of NNs to achieve higher performance. However, such performance-centric developments in artificial intelligence have significant challenges. First of all, the NNs from performance-centric developments are unavailable for individuals and small businesses and make them dependent on a few major companies, called Big Techs. Individuals and small businesses, despite having novel technologies for applications, fail to use NNs efficiently if they do not have access to the large resources of Big Techs. Second, the performance improvement of NNs based on large-scale computational and storage resources is impractical in non-ideal real-world environments. This is because these performance-centric developments are studied under the assumption that NNs are trained in an ideal environment equipped with substantial computational resources. In real-world environments, there are several communication conditions and distributed computing resources are generally utilized. In addition, non-independent and identically distributed (non-IID) datasets are utilized in real-world environments. Lastly, the complexity of data is increasing even more quickly and current performance-centric development cannot fundamentally reduce the computational complexity and demands substantial costs. To cope with these challenges of performance-centric NN developments, this paper proposes a novel direction for the development of the next-generation NNs. First of all, this paper proposes federated learning with slimmable neural network (SNN), named SlimFL, for non-IID datasets. Our SlimFL replace NNs with SNNs to maximize the performance of federated learning in various communication environments. Second, this paper proposes scalable 3D quantum convloutional neural network (sQCNN-3D) to efficiently analyze the complex 3D point cloud data. With intrinsic natures of quantum computing that obtains multiple information using a single computing unit (i.e., superposition) and convolutes several information in parallel (i.e., entanglement), our sQCNN-3D shows great performance in point classification tasks. In addition, this paper proposes an efficient training algorithm for sQCNN-3D that varies the fidelity between convolutional filters for achieving various features. Third, this paper proposes multi resolution quantum convolution neural network (MR-QCNN) that adopts quantum convolutional neural network in federated learning framework. To reduce the domain gap between MR-QCNNs in heterogeneous resolutions, this paper utilizes knowledge distillation. With these characteristics, our MR-QCNN outperforms other NNs in various datasets and shows the smallest domain gap. Lastly, this paper proposes quantum convolution based object detection (QCOD) using a faster quantum convolution that re-uploads and re-constructs the channel information to cope with speed constraints of the object detection. By using our faster quantum convolution, our QCOD shows great improvement in speed and shows feasible performance. Furthermore, this paper proposes heterogeneous knowledge distillation that transfers the knowledge of pre-trained classical object detection to our QCOD to improve the performance. Through each chapter, this paper corroborates the feasibility and performance of SNNs and QCNNs as the next-generation NNs via various experiments and analyses.

목차 (Table of Contents)

• 1 Introduction 1
• 1.1 Core Algorithms and Architectures 3
• 1.2 Motivations and Contributions 6
• 1.3 Organization 7
• 1.4 Outline and Contributions 9
• 1.4.1 Stereoscopic Scalable Quantum Convolutional Neural Networks 9
• 1.4.2 Joint Federated Learning and Knowledge Distillation for Multi-Resolution 3D Quantum Convolutional Neural Networks 10
• 1.4.3 Joint Superposition Coding and Training for Federated Learning over Multi-Width Neural Networks 11
• 1.4.4 Fast Quantum Convolution Neural Networks for Low-Complexity Object Detection in Autonomous Driving Applications 12
• 2 Stereoscopic Scalable Quantum Convolutional Neural Networks 13
• 2.1 Introduction 13
• 2.2 Preliminaries 16
• 2.2.1 Point Cloud Processing with Classical CNN 16
• 2.2.2 Quantum CNN 17
• 2.2.3 Motivation 20
• 2.2.4 sQCNN-3D Architecture 21
• 2.3 3D Scalable Quantum Convolutional Neural Networks (sQCNN-3D) for Point Cloud Processing 24
• 2.3.1 RF-Train 24
• 2.3.2 sQCNN-3D Overall Algorithmic Procedure 25
• 2.4 Performance Evaluation 28
• 2.4.1 Experimental Setting 28
• 2.4.2 Experimental Results 30
• 2.5 Conclusions and Future Work 32
• 3 Joint Federated Learning and Knowledge Distillation for Multi-Resolution 3D Quantum Convolutional Neural Networks 35
• 3.1 Introduction 35
• 3.2 Quantum Machine Learning – A Primer 38
• 3.2.1 Basic Quantum Operations 38
• 3.2.2 Quantum Neural Networks 41
• 3.3 Multi-Resolution QCNN 42
• 3.3.1 Motivation 42
• 3.3.2 Architecture and its Process 43
• 3.4 MR-QCNN Training to Handling Domain Gap over Multi-Resolution 45
• 3.4.1 Overview 45
• 3.4.2 Local Training with Knowledge Distillation 46
• 3.4.3 MR-QCNN Aggregation 47
• 3.5 Convergence Analysis 48
• 3.5.1 Setup 48
• 3.5.2 Convergence Analysis 49
• 3.6 Numerical Experiments 52
• 3.6.1 Experiment Setting 52
• 3.6.2 Numerical Results 55
• 3.7 Discussions 57
• 3.8 Related Work 59
• 3.9 Concluding Remarks 60
• 4 Joint Superposition Coding and Training for Federated Learning over Multi-Width Neural Networks 63
• 4.1 Introduction 63
• 4.2 Related Work 67
• 4.2.1 Multi-Width/Depth Neural Networks 67
• 4.2.2 Superposition Coding & Successive Decoding 67
• 4.2.3 FL Convergence Analysis 68
• 4.3 Local Model Architecture and Training 69
• 4.3.1 Ultra Light SNN Architecture 69
• 4.3.2 Superposition SNN Training 70
• 4.4 Global Model Aggregation with Superposition Coding & Successive Decoding 73
• 4.4.1 Superposition Coding & Successive Decoding 73
• 4.4.2 SlimFL Operations 76
• 4.5 SlimFL Convergence Analysis 77
• 4.6 Experiments 83
• 4.6.1 Experimental Setup 83
• 4.6.2 Guidelines for SlimFL 85
• 4.6.3 Performance of SlimFL 88
• 4.6.4 Communication and Energy Efficiency 90
• 4.7 Conclusion 92
• 5 Fast Quantum Convolution Neural Networks for Low-Complexity Object Detection in Autonomous Driving Applications 94
• 5.1 Introduction 94
• 5.2 Related Work 98
• 5.3 Quantum Machine Learning 100
• 5.4 Fast Quantum Convolutional Neural Networks 101
• 5.4.1 Motivation 102
• 5.4.2 Architecture of Fast Quantum Convolutional Neural Network 103
• 5.4.3 Strategy of Fast Quantum Convolution 106
• 5.5 Quantum Object Detection 110
• 5.5.1 Motivation of Quantum Region Proposal Network 110
• 5.5.2 Architecture of Quantum Region Proposal Network 110
• 5.5.3 Heterogeneous Knowledge Distillation Training 111
• 5.5.4 Loss of QRPN 112
• 5.5.5 Implementation Details 112
• 5.6 Performance Evaluation 113
• 5.6.1 Setup 114
• 5.6.2 Evaluation Results 115
• 5.7 Concluding Remarks 117
• 6 Concluding Remarks 122
• Reference 124
• Appendix 146
• A Appendix of Chapter 1 146
• A.1 Local SNN Training Algorithm 146

참고문헌 (Reference)

1. Broadcast channels,, T. Cover, IEEE Transactions on Information Theory, vol. 18, no. 1, pp. 2–14, , 1972

2. Attention is all you need, L. Jones, J. Uszkoreit, N. Parmar, L. Kaiser and, I. Polosukhin, N. Shazeer, A. N. Gomez, A. Vaswani, in Proc. of the Advances in Neural Information Processing Systems (Neurips), Long Beach, CA, USA pp. 5998–6008, , 2017

3. Slimmable neural networks, L. Yang, N. Xu, J. Yu, J. Yang and, T. Huang, in Proc. of International Conference on Learning Representation (ICLR), New Orleans, LA, USA, , 2019

4. Agnostic federated learning, G. Sivek and, M. Mohri, A. T. Suresh, in Proc. International Conference on Machine Learning (ICML). Long Beach, CA, USA: PMLR, pp. 4615–4625, , 2019

5. Quantum information theory,, P. W. Shor, C. H. Bennett and, IEEE Transactions on Information Theory, vol. 44, no. 6, pp. 2724–2742, , 1998

6. Spatio-temporal split learning, J. Kim, S. Yoo, S. Park, S. Jung and, in Proc IEEE/IFIP International Conference on Dependable Systems and Networks (DSN), Taipei, Taiwan, , 2021

7. A survey on multi-task learning, Y. Zhang and, Q. Yang, IEEE Transactions on Knowledge and Data Engineering, vol. 34, no. 12, , 2022

8. Quantum neural network compression, Y. Wang andW. Jiang, Z. Hu, Y. Lin, P. Dong, Z. Wang, in Proc. of IEEE/ACM International Conference on Computer- Aided Design (ICCAD) 2022, San Diego, California, USA p. 140, , 2022

9. Relational knowledge distillation,, Y. Lu and, M. Cho, D. Kim, W. Park, in Proc. of IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA pp. 3967–3976, , 2019

10. Entanglement induced barren plateaus,, N. Wiebe, M. Kieferov´a and, C. O. Marrero, PRX Quantum, vol. 2, no. 4, p. 040316, , 2021

11. Quantum convolutional neural networks, I. Cong, S. Choi and, M. D. Lukin, vol. 15, no. 12, pp. 1273–1278, , 2019

12. Fundamentals ofWireless Communications, D. N. C. Tse and, P. Viswanath, Cambridge University Press, , 2005

13. Quantum-inspired support vector machine,, H.-L. Huang, C. Ding, T.-Y. Bao and, IEEE Transactions on Neural Networks and Learning Systems, pp. 1–13, , 2021

14. Vision meets robotics: The KITTI dataset, C. Stiller and, R. Urtasun, A. Geiger, P. Lenz, vol. 32, no. 11, pp. 1231– 1237, , 2013

15. Elementary gates for quantum computation,, A. Barenco, H. Weinfurter, T. Sleator, N. Margolus, P. Shor, R. Cleve, D. P. DiVincenzo, C. H. Bennett, J. A. Smolin and, Physical review A, vol. 52, no. 5, p. 3457, , 1995

16. Learning to summarize with human feedback, P. F. Christiano, D. M. Ziegler, R. Lowe, A. Radford, N. Stiennon, J. Wu, L. Ouyang, D. Amodei and, C. Voss, in Proc. of the Annual Conference on Neural Information Processing Systems (NeurIPS) 2020, Virtual, , 2020

17. The elastic lottery ticket hypothesis, in, Y. Cheng, Z. Gan, J. Liu and, Z. Wang, X. Chen, S. Wang, Advances in Neural Information Processing Systems (NeurIPS), Virtual, pp. 26 609–26 621, , 2021

18. Blockchained on-device federated learning,, M. Bennis and, J. Park, S.-L. Kim, H. Kim, vol. 24, no. 6, pp. 1279–1283, , 2019

19. Wireless network intelligence at the edge,, J. Park, S. Samarakoon, M. Debbah, M. Bennis and, Proceedings of the IEEE, vol. 107, no. 11, pp. 2204–2239, , 2019

20. Deep learning for 3D point clouds: A survey,, H. Wang, Y. Guo, H. Liu, Q. Hu, M. Bennamoun, L. Liu and, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 43, no. 12, pp. 4338–4364, , 2020

21. On the convergence of FedAvg on non-iid data, K. Huang, W. Yang, Z. Zhang, X. Li, S. Wang and, in Proc. of International Conference on Learning Representation (ICLR), Addis Ababa, Ethiopia, , 2020

22. Channel estimation for wireless OFDM systems,, H. Arslan, M. K. Ozdemir and, IEEE Communications Surveys & Tutorials, vol. 9, no. 2, pp. 18–48, , 2007

23. Distilling the knowledge in a neural network,, J. Dean, G. E. Hinton, O. Vinyals and, in Proc. of the Conference on Neural Information Processing Systems (NIPS) Deep Learning and Representation Learning Workshop, pp. 1–9, , 2015

24. Robust data encodings for quantum classifiers,, B. Coyle, R. LaRose and, Physical Review A, vol. 102, no. 3, p. 32420, , 2020

25. Scalable quantum convolutional neural networks, H. Baek, W. J. Yun and, J. Kim, CoRR, vol. abs/2209.12372, , 2022

26. Dynamic graph cnn for learning on point clouds,, J. M. Solomon, Y. Wang, S. E. Sarma, Z. Liu, Y. Sun, M. M. Bronstein and, ACM Transactions on Graphics, vol. 38, no. 5, pp. 146:1–146:12, , 2019

27. Knowledge distillation beyond model compression, F. Sarfraz, E. Arani and, B. Zonooz, in Proc. of the IEEE International Conference on Pattern Recognition (ICPR), Milan, Italy pp. 6136–6143, , 2020

28. An introduction to convolutional neural networks, R. Nash, K. O’Shea and, CoRR, vol. abs/1511.08458, , 2015

29. Learning to learn variational quantum algorithm,, Q. Xu, X. Tan and, R. Huang, IEEE Transactions on Neural Networks and Learning Systems, pp. 1–11, , 2022

30. Local SGD converges fast and communicates little, S. U. Stich, in Proc of International Conference on Learning Representations (ICLR), Vancouver, BC, Canada, , 2018

31. Quantum machine learning in high energy physics,, A. Pesah, M. Schuld, J. Vlimant, S. Vallecorsa and, K. Terashi, G. N. Perdue, W. Guan, Machine Learning: Science Technology, vol. 2, no. 1, p. 11003,, , 2021

32. Quantum multi-agent meta reinforcement learning,, W. J. Yun, J. Park and, J. Kim, CoRR, vol. abs/2208.11510, , 2022

33. Advances and open problems in federated learning,, H. B. McMahan, Z. Charles, B. Avent, G. Cormode, A. N. Bhagoji, A. Bellet, K. Bonawitz, R. Cummings et al., P. Kairouz, M. Bennis, Foundations and Trends® in Machine Learning, vol. 14, no. 1–2, pp. 1–210, 2021., , 2021

34. Shapenet: An information-rich 3d model repository, A. X. Chang, T. A. Funkhouser, Q. Huang, L. J. Guibas, Z. Li, S. Savarese, J. Xiao, H. Su, F. Yu, P. Hanrahan, L. Yi and, S. Song, M. Savva, CoRR, vol. abs/1512.03012, 2015., , 2015

35. Quantum Machine Learning in Feature Hilbert Spaces, M. Schuld and, N. Killoran, vol. 122, p. 040504, , 2019

36. Second: Sparsely embedded convolutional detection,, Y. Yan, Y. Mao and, B. Li, vol. 18, no. 10, p. 3337, , 2018

37. Big data deep learning: Challenges and perspectives, X. Lin, X. Chen and, IEEE Access, vol. 2, pp. 514–525, , 2014

38. QFCNN: Quantum Fourier convolutional neural network, F. Shen and, J. Liu, CoRR, vol. abs/2106.10421, , 2021

39. Quantum federated learning with decentralized data,, Q. Xu, R. Huang, X. Tan and, IEEE Journal of Selected Topics in Quantum Electronics, vol. 28, no. 4: Machine Learning in Photonic Communication and Measurement Systems, pp. 1–10, , 2022

40. Torchquantum case study for robust quantum circuits, Z. Li, S. Han, W. Jiang, Z. Liang, F. T. Chong and, D. Z. Pan, J. Gu, Y. Ding, H. Wang, Y. Shi, in Proc. of IEEE/ACM International Conference on Computer-Aided Design (ICCAD), San Diego, California, USA p. 136, , 2022

41. 17LiDAR market size & growth forecasts (2022–2030), L. Smriti, W. Preeti and, Online Available https://www. gminsights. com/industry-analysis/lidar-l ight-detection-and-ranging-market, , 2021

42. Data re-uploading for a universal quantum classifier, J. I. Latorre, A. Cervera-Lierta, E. Gil-Fuster and, A. P´erez-Salinas, Verein zur Forderung des Open Access Quantum, , 2020

43. Depth-controllable very deep superresolution network, J. Kim, T.-H. Kim, D. Kim, J. Kwon and, in Proc. of IEEE International Joint Conference on Neural Networks (IJCNN), Budapest, Hungary, , 2019

44. General parameter-shift rules for quantum gradients,, J. Izaac, D. Wierichs, C. Y. Lin, C. Wang and, vol. 6, p. 677, , 2022

45. PointGrid: A deep network for 3D shape understanding, Y. Duan, T. Le and, in Proc IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, UT, USA pp. 9204–9214, , 2018

46. A fast quantum mechanical algorithm for database search, L. K. Grover, in Proc ACM Symposium on the Theory of Computing (STOC), Philadelphia, PA, USA pp. 212–219, , 1996

47. Measuring the algorithmic efficiency of neural networks, D. Hernandez and, T. B. Brown, arXiv preprint, abs/2005.04305, , 2020

48. Supervised learning with quantumenhanced feature spaces, A. D. C´orcoles, A. W. Harrow, J. M. Gambetta, A. Kandala, J. M. Chow and, V. Havl´ıˇcek, K. Temme, vol. 567, no. 7747, pp. 209–212, , 2019

49. Variational quantum policies for reinforcement learning, C. Gyurik, V. Dunjko, S. Marshall, S. Jerbi, H. J. Briegel and, in Proc. of Advances in Neural Information Processing Systems (NeurIPS), Virtual, , 2021

50. Pruning deep neural networks from a sparsity perspective, J. Ding and, G. Wang, E. Diao, J. Zhang, V. Tarokh, Y. Yang, in Proc. of the International Conference on Learning Representations (ICLR), Kigali, Rwanda, , 2023

51. 3d shapenets: A deep representation for volumetric shapes, A. Khosla, S. Song, F. Yu, L. Zhang, X. Tang and, Z. Wu, J. Xiao, in Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1912–1920, , 2015

52. An introduction to quantum machine learning for engineers, O. Simeone, CoRR, vol. abs/2205.09510, , 2022

53. CNN based region proposals for efficient object detection, A. K. Roy-Chowdhury, J. H. Bappy and, in Proc. of the IEEE International Conference on Image Processing (ICIP), Phoenix, AZ, USA pp. 3658–3662, , 2016

54. Challenges and opportunities in quantum machine learning,, H. Huang, P. J. Coles, L. Cincio and, M. Cerezo, G. Verdon, Nature Computational Science, vol. 2, no. 9, pp. 567–576, , 2022

55. Data collection and quality challenges for deep learning,, J. Lee, S. Whang and, Proceedings of the VLDB Endowment, vol. 13, no. 12, pp. 3429–3432, , 2020

56. Quantum algorithms for deep convolutional neural networks, A. Prakash, I. Kerenidis, J. Landman and, in Proc. of the International Conference on Learning Representations (ICLR), Addis Ababa, Ethiopia, , 2020

57. Simultaneous training of partially masked neural networks, M. Jaggi and, A. Mohtashami, S. U. Stich, arxiv preprint, abs/2106.08895, , 2021

58. A tutorial on quantum convolutional neural networks (QCNN), J. Choi and, S. Oh, J. Kim, in Proc IEEE International Conference on Information and Communication Technology Convergence (ICTC), pp. 236–239, , 2020

59. Entanglement properties of random invariant quantum states,, W. Li, W. Xie and, vol. 22, no. 11&12, pp. 901–923, , 2022

60. Quantum support vector machine for big data classification,, M. Mohseni and, P. Rebentrost, S. Lloyd, vol. 113, no. 13, p. 130503, , 2014

61. Parallel gradient distribution in unconstrained optimization, L. Mangasarian, vol. 33, no. 6, pp. 1916–1925, , 1995

62. Stereoscopic scalable quantum convolutional neural networks,, J. Kim, W. J. Yun, H. Baek, S. Park and, Neural Networks, vol. 165, pp. 860–867, , 2023

63. Avoiding barren plateaus with classical deep neural networks,, L. Friedrich and, J. Maziero, vol. 106, no. 4, p. 042433, , 2022

64. Joint rate and power allocation for NOMA with statistical CSI, J. Choi, IEEE Trans. Communi., vol. 65, no. 10, pp. 4519–4528, , 2017

65. Quantumnas: Noise-adaptive search for robust quantum circuits, D. Z. Pan, J. Gu, Y. Lin, Y. Ding, H. Wang, F. T. Chong and, S. Han, in Proc. of the IEEE International Symposium on High-Performance Computer Architecture (HPCA), Seoul, South Korea pp. 692–708, , 2022

66. Semi-supervised image deraining using knowledge distillation,, D. Ren, X. Cui, Y. Chen and, P. Zhu, C. Wang, IEEE Transactions on Circuits and Systems and Video Technology, vol. 32, no. 12, pp. 8327–8341, , 2022

67. Barren plateaus in quantum neural network training landscapes,, S. Boixo, H. Neven, J. R. McClean, V. N. Smelyanskiy, R. Babbush and, CoRR, vol. abs/1803.11173, , 2018

68. Hierarchical textconditional image generation with CLIP latents, C. Chu and, A. Nichol, A. Ramesh, P. Dhariwal, M. Chen, CoRR, vol. abs/2204.06125, , 2022

69. Human-machine shared driving: Challenges and future directions,, S. Ansari, F. Naghdy and, H. Du, IEEE Transactions on Intelligent Vehicles, vol. 7, no. 3, pp. 499–519, , 2022

70. Universally slimmable networks and improved training techniques, J. Yu and, T. S. Huang, in Proc of IEEE/CVF International Conference on Computer Vision (ICCV), Seoul, South Korea pp. 1803–1811, , 2019

71. Quantum supremacy using a programmable superconducting processor, F. G. Brandao, R. Biswas, R. Barends, R. Babbush, F. Arute, S. Boixo, K. Arya, D. Bacon, D. A. Buell et al., J. C. Bardin, vol. 574, no. 7779, pp. 505–510, , 2019

72. Tighter theory for local SGD on identical and heterogeneous data, A. Khaled, K. Mishchenko and, P. Richt´arik, in Proc. of the International Conference on Artificial Intelligence and Statistics (AISTATS), vol. 108, Virtual, 2020, pp. 4519– 4529, , 2020

73. CNN-RNN: A unified framework for multi-label image classification, J. Mao, C. Huang and, Y. Yang, J. Wang, Z. Huang, W. Xu, in Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA pp. 2285–2294, , 2016

74. Efficient quantum image classification using single qubit encoding, P. Easom-McCaldin, A. Bouridane, S. Al-Maadeed, A. Belatreche, R. Jiang and, IEEE Transactions on Neural Networks and Learning Systems, pp. 1–15, , 2022

75. Universal approximation theorem for equivariant maps by group CNNs, A. Sannai, W. Kumagai and, CoRR, vol. abs/2012.13882, , 2020

76. Very deep convolutional networks for large-scale image recognition, K. Simonyan and, A. Zisserman, in Proc. of the International Conference on Learning Representations (ICLR), San Diego, CA, USA, , 2015

77. Algorithms for quantum computationdiscrete logarithms and factoring, P. W. Shor, in Proc IEEE Symposium on Foundations of Computer Science (FOCS), Santa Fe, NM, USA pp. 124–134, , 1994

78. Provably efficient machine learning for quantum many-body problems,, V. V. Albert and, G. Torlai, H. Huang, R. Kueng, J. Preskill, Science, vol. 377, no. 6613, p. eabk3333, , 2022

79. Quantum algorithms for supervised and unsupervised machine learning, S. Lloyd, P. Rebentrost, M. Mohseni and, CoRR, vol. abs/1307.0411, , 2013

80. Temperature annealing knowledge distillation from averaged teacher,, Z. Zhang and, T. Luo, X. Gu, in Proc. of IEEE International Conference on Distributed Computing Systems (ICDCS) Workshops, Bologna, Italy pp. 133–138, , 2022

81. Absence of barren plateaus in quantum convolutional neural networks,, S. Wang, P. J. Coles, A. T. Sornborger and, T. Volkoff, M. Cerezo, A. Pesah, CoRR, vol. abs/2011.02966, , 2020

82. Large-scale point cloud semantic segmentation with superpoint graphs, L. Landrieu and, M. Simonovsky, in Proc IEEE Conference on Computer Vision and Pattern Recognition, (CVPR), Salt Lake City, UT, USA pp. 4558–4567, , 2018

83. QuEst: Graph transformer for quantum circuit reliability estimation,, W. Jiang, Y. Ding, X. Qian et al., P. Liu, Z. Li, Z. Liang, J. Cheng, H. Wang, Y. Shi, J. Gu, arXiv preprint arXiv:2210.16724, , 2022

84. Application of non-orthogonal multiple access in lte and 5g networks,, Q. Sun, H. V. Poor, I. Chih-Lin and, M. Elkashlan, J. Choi, Z. Ding, Y. Liu, vol. 55, no. 2, pp. 185–191, , 2017

85. Quantum machine learning for material synthesis and hardware security, C. Beaudoin, S. Kundu, S. Ghosh, R. O. Topaloglu and, in Proc of the IEEE/ACM International Conference on Computer-Aided Design (ICCAD), San Diego, California, USA p. 120, , 2022

86. Temporal self-ensembling teacher for semi-supervised object detection,, L. Liu and, Y. Guo, Y. Tian, K. Cao, C. Chen, S. Dong, IEEE Transactions on Multimedia, vol. 24, pp. 3679–3692, , 2022

87. QOC: Quantum on-chip training with parameter shift and gradient pruning, D. Z. Pan and, S. Han, J. Gu, Z. Li, H. Wang, Y. Ding, in Proc. of ACM/IEEE Design Automation Conference (DAC), San Francisco, CA, USA pp. 665–660, , 2022

88. Voxelnet: End-to-end learning for point cloud based 3d object detection, Y. Zhou and, O. Tuzel, in Proc IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, UT, USA pp. 4490–4499, , 2018

89. Semantic3d. net: A new large-scale point cloud classification benchmark,, N. Savinov, T. Hackel, L. Ladicky, M. Pollefeys, J. D. Wegner, K. Schindler and, CoRR, vol. abs/1704.03847, , 2017

90. A LiDAR point cloud generator: from a virtual world to autonomous driving, A., L., S. Sanjit, Y. Xiangyu, W. Bichen, S. Alberto, K. Kurt and, in Proc. ACM International Conference on Multimedia Retrieval (ICMR), Yokohama, Japan pp. 458–464, , 2018

91. Communication-efficient learning of deep networks from decentralized data, D. Ramage, S. Hampson and, E. Moore, B. A. y Arcas, H. B. McMahan, in Proc. of International Conference on Artificial Intelligence and Statistics (AISTATS), Fort Lauderdale, FL, USA pp. 1273–1282, , 2017

92. Efficient evaluation of quantum observables using entangled measurements,, I. Hamamura and, T. Imamichi, vol. 6, no. 1, pp. 1–8, , 2020

93. IBM Quantum’s mission is to bring useful quantum computing to the world, J. Gambetta, IBM Quantum Summit. IBM Quantum, , 2022

94. Quantum machine learning of large datasets using randomized measurements,, C. N. Self and, T. Haug, M. Kim, Machine Learning: Science and Technology, 2023, , 2023

95. VACSEN: A visualization approach for noise awareness in quantum computing, S. Ruan, Q. Guan, W. Jiang, Y. Mao and, Y. Wang, IEEE Transactions on Visualization and Computer Graphics, vol. 29, no. 1, pp. 462–472, , 2022

96. A review of object detection models based on convolutional neural network,, A. Sufian and, F. Sultana, P. Dutta, Intelligent computing: image processing based applications, pp. 1–16, , 2020

97. Parallelizing the queries in a bucketbrigade quantum random access memory,, O. Oumarou and, R. Basmadjian, A. Paler, Physical Review A, vol. 102, no. 3, p. 032608, , 2020

98. VoxNet: A 3D convolutional neural network for real-time object recognition, D. Maturana and, S. A. Scherer, in Proc IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Hamburg, Germany pp. 922–928, , 2015

99. Learning anytime predictions in neural networks via adaptive loss balancing, H. Hu, M. Hebert and, D. Dey, J. A. Bagnell, in Proc. of AAAI Conference on Artificial Intelligence (AAAI), Honolulu, Hawaii, USA pp. 3812–3821, , 2019

100. Point cloud data conversion into solid models via point-based voxelization,, T. Hinks, H. Carr, L. Truong-Hong et al., vol. 139, no. 2, pp. 72–83, , 2013

101. Pointnet: Deep learning on point sets for 3d classification and segmentation, K. Mo and, C. R. Qi, L. J. Guibas, H. Su, in Proc IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, HI, USA pp. 652– 660, , 2017

102. Quantum convolutional neural networks for high energy physics data analysis,, S. Y. Chen, S. Yoo, H. Yu and, T. Wei, C. Zhang, CoRR, vol. abs/2012.12177, , 2020

103. Advances in the quantum theoretical approach to image processing applications,, N. Aburaed, H. Bhaskar, F. S. Khan and, ACM Computing Surveys, vol. 49, no. 4, pp. 75:1–75:49, , 2017

104. Faster R-CNN: Towards real-time object detection with region proposal networks,, K. He, J. Sun, S. Ren, R. B. Girshick and, IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 39, no. 6, pp. 1137–1149, , 2017

105. Hardware approximate techniques for deep neural network accelerators: A survey,, J. Henkel, G. Zervakis, D. Soudris and, G. Armeniakos, ACM Comput. Surv., vol. 55, no. 4, pp. 83:1–83:36, 2023, , 2023

106. Uses and abuses of the cross-entropy loss: Case studies in modern deep learning, G. Loaiza-Ganem, E. Gordon-Rodr´ıguez, J. P. Cunningham, G. Pleiss and, CoRR, vol. abs/2011.05231, , 2020

107. On barren plateaus and cost function locality in variational quantum algorithms,, A. Uvarov and, J. D. Biamonte, Journal of Physics A: Mathematical and Theoretical, vol. 54, no. 24, p. 245301, , 2021

108. Federated learning over wireless networks: Optimization model design and analysis, W. Bao, A. Y. Zomaya, N. H. Tran, C. S. Hong, M. N. H. Nguyen and, in Proc. of IEEE Conference on Computer Communications (INFOCOM), pp. 1387–1395, , 2019

109. PointDAN: A multi-scale 3D domain adaption network for point cloud representation, C. J. Kuo and, Y. Fu, C. Qin, H. You, L. Wang, in Proc. Advances in Neural Information Processing Systems (NeurIPS), Vancouver, BC, Canada pp. 7190–7201, , 2019

110. Quanvolutional neural networks: Powering image recognition with quantum circuits,, S. Shakya, M. Henderson, T. Cook, S. Pradhan and, vol. 2, no. 1, pp. 1–9, , 2020

111. Quantum multi-agent reinforcement learning via variational quantum circuit design,, W. J. Yun, J. P. Kim, H. Cho, Y. Kwak, J. Kim, J. Park and, S. Jung, in Proc IEEE International Conference on Distributed Computing Systems (ICDCS), Bologna, Italy, , 2022

112. Resmlp: Feedforward networks for image classification with data-efficient training, A. Joulin, M. Cord, M. Caron, J. Verbeek and, E. Grave, A. El-Nouby, H. Touvron, G. Izacard, P. Bojanowski, H. J´egou, G. Synnaeve, IEEE Trans. Pattern Anal. Mach. Intell., vol. 45, no. 4, pp. 5314–5321, 2023., , 2023

113. Boost 3-D object detection via point clouds segmentation and fused 3-D GIoU-L, loss,, H. Li, Y. Chen, R. Gao and, D. Zhao, IEEE Transactions on Neural Networks and Learning Systems, vol. 33, no. 2, pp. 762–773, , 2022

114. Learning graph convolutional networks based on quantum vertex information propagation, L. Bai, E. R. Hancock, L. Rossi, L. Cui, P. S. Yu and, Y. Wang, Y. Jiao, in Proc. IEEE International Conference on Data Engineering, (ICDE), Kuala Lumpur, Malaysia pp. 3132–3133, , 2022

115. Pooling acceleration in the DaVinci architecture using Im2col and Col2im instructions, G. Colmenares and, C. S. Rohwedder, J. P. L. de Carvalho, J. N. Amaral, G. Ara´ujo, K. A. Wang, in Proc. of the IEEE International Parallel and Distributed Processing Symposium (IPDPS), Portland, OR, USA, , 2021

116. Training hybrid classical-quantum classifiers via stochastic variational optimization, I. Nikoloska and, O. Simeone, IEEE Signal Processing Letters, vol. 29, pp. 977–981, , 2022

117. Projection valued measure-based quantum machine learning for multi-class classification, J. Kim, H. Baek and, W. J. Yun, CoRR, vol. abs/2210.16731, , 2022

118. Quantum federated learning with entanglement controlled circuits and superposition coding, H. Baek, J. Park, S. Jung, M. Bennis and, W. J. Yun, J. Kim, J. P. Kim, CoRR, vol. abs/2212.01732, , 2022

119. Classically approximating variational quantum machine learning with random Fourier features, S. Thabet, E. Kashefi, J. Landman, C. Dalyac, H. Mhiri and, in Proc. of the International Conference on Learning Representations (ICLR), Kigali, Rwanda, , 2023

120. Joint distribution alignment via adversarial learning for domain adaptive object detection,, B. Zhang, T. Chen, R. Li, B. Wang and, IEEE Transactions on Multimedia, vol. 24, pp. 4102–4112, , 2022

121. Measuring the effects of non-identical data distribution for federated visual classification, M. Brown, T. H. Hsu, H. Qi and, Conference on Neural Information Processing Systems (NeurIPS) Workshop on Federated Learning for Data Privacy and Confidentiality (available on arXiv preprint, abs/1909.06335), , 2019

122. Dual-frequency quantum phase estimation mitigates the spectral leakage of quantum algorithms,, L. Hanzo, S. X. Ng, G. Long and, Y. Xiong, IEEE Signal Processing Letters, vol. 29, pp. 1222–1226, , 2022

123. Effect of data encoding on the expressive power of variational quantum-machine-learning models, J. J. Meyer, R. Sweke and, M. Schuld, Physical Review A, vol. 103, no. 3, p. 032430, , 2021

124. Deeper look at image salient object detection: Bi-stream network with a small training dataset,, C. Chen, A. Hao and, S. Li, Z. Wu, H. Qin, IEEE Transactions on Multimedia, vol. 24, pp. 73–86, , 2022

125. Joint superposition coding and training for federated learning over multi-width neural networks,, J. Park and, Y. Kwak, S. Jung, H. Baek, M. Bennis, M. Ji, J. Kim, W. J. Yun, in Proc. of the IEEE Conference on Computer Communications (INFOCOM). London, United Kingdom: IEEE, pp. 1729–1738, , 2022

126. Quantum- NAT: quantum noise-aware training with noise injection, quantization and nor- malization, D. Z. Pan and, H. Wang, S. Han, F. T. Chong, J. Gu, Z. Li, Y. Ding, in Proc. of ACM/IEEE Design Automation Conference (DAC), San Francisco, CA, USA, pp. 1–6, , 2022

127. Hierarchical deep reinforcement learning-based propofol infusion assistant framework in anesthesia,, D. Mohaisen, W. J. Yun, M. Shin, K. Lee and, J. Kim, IEEE Transactions on Neural Networks and Learning Systems, pp. 1–12, , 2022

128. Measuring analytic gradients of general quantum evolution with the stochastic parameter shift rule,, G. E. Crooks, L. Banchi and, vol. 5, p. 386, , 2021

129. Communication-efficient and distributed learning over wireless networks: Principles and applications,, S. Samarakoon, M. Debbah, A. Elgabli, J. Park, M. Bennis, J. Kim, S.-L. Kim and, Proceedings of the IEEE, vol. 109, no. 5, pp. 796–819, , 2021

130. Communication-efficient and distributed learning over wireless networks: Principles and applications,, J. Kim, M. Debbah, A. Elgabli, M. Bennis, J. Park, S. Kim and, S. Samarakoon, Proceedings of the IEEE, vol. 109, no. 5, pp. 796–819, , 2021

131. A review of quantum key distribution protocols in the perspective of smart grid communication security,, P. Kong, IEEE Systems Journal, vol. 16, no. 1, pp. 41–54, , 2022

132. Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding, H. Mao and, S. Han, W. J. Dally, in Proc. of International Conference on Learning Representations (ICLR), San Juan, Puerto Rico, , 2016

133. Can noise on qubits be learned in quantum neural network? A case study on quantumflow (invited paper), in, Z. Wang, Y. Shi and, L. Yang, J. Yang, Z. Liang, W. Jiang, IEEE/ACM International Conference On Computer Aided Design (ICCAD), Munich, Germany pp. 1–7, , 2021

134. CAA-Net: Conditional atrous CNNs with attention for explainable device-robust acoustic scene classification, B. W. Schuller, J. Han, M. D. Plumbley and, Q. Kong, Z. Ren, IEEE Transactions on Multimedia, vol. 23, pp. 4131–4142, , 2021

135. 3D scalable quantum convolutional neural networks for point cloud data processing in classification applications, J. Kim, W. J. Yun and, H. Baek, in Proc of AAAI Workshop on AI to Accelerate Science and Engineering (AI2ASE), Washington, DC, USA, , 2023

136. Quantum tomography of entangled qubits by time-resolved single-photon counting with time-continuous measurements,, A. Czerwinski, vol. 21, no. 9, p. 332, , 2022

137. Model soups: averaging weights of multiple fine-tuned models improves accuracy with- out increasing inference time in, R. G. Lopes, L. Schmidt, H. Namkoong, G. Ilharco, A. Farhadi, R. Roelofs, A. S. Morcos, Y. Carmon, S. Kornblith and, M. Wortsman, S. Y. Gadre, Proc. of the International Conference on Machine Learning, (ICML), vol. 162, Baltimore, Maryland, USA, July 2022, pp. 23 965–23 998., , 2022

138. FedZKT: Zero-shot knowledge transfer towards resource-constrained federated learning with heterogeneous on-device models, L. Zhang, X. Yuan, D. Wu and, in Proc. of IEEE International Conference on Distributed Computing Systems (ICDCS), Bologna, Italy, pp. 928–938, , 2022

139. FedDG: Federated domain generalization on medical image segmentation via episodic learning in continu- ous frequency space, C. Chen, Q. Liu, Q. Dou and, P.-A. Heng, J. Qin, in Proc. of IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA, pp. 1013–1023, , 2021

140. Quantum convolutional neural network for resource-efficient image classification: A quantum random access memory (QRAM) approach,, J. Choi, J. Kim and, J. Kim, S. Oh, in Proc. of the IEEE International Conference on Information Networking (ICOIN), Jeju Island, South Korea, pp. 50–52, , 2021

141. Level-K reasoning, deep reinforcement learning, and Monte Carlo decision process for fast and safe automated lane change and speed management,, A. Vahidi, A. Karimi and, S. Karimi, IEEE Transactions on Intelligent Vehicles, vol. 8, no. 6, pp. 3556–3571, 2023, , 2023