Daily Arxiv

This is a page that curates AI-related papers published worldwide.
All content here is summarized using Google Gemini and operated on a non-profit basis.
Copyright for each paper belongs to the authors and their institutions; please make sure to credit the source when sharing.

CTRQNets & LQNets: Continuous Time Recurrent and Liquid Quantum Neural Networks

Created by
  • Haebom

Author

Alejandro Antonio Mayorga, Alexander Yuan, Andrew Yuan, Tyler Wooldridge, Xiaodi Wang

Outline

To overcome the limitations of conventional quantum neural networks with static structures, this paper proposes a liquid quantum neural network (LQNet) with dynamic intelligence and a continuous-time recurrent quantum neural network (CTRQNet). Both models demonstrate significant accuracy improvements over conventional quantum neural networks, achieving up to a 40% improvement in accuracy on the CIFAR-10 binary classification task. This suggests their potential contribution to understanding the black box of quantum machine learning.

Takeaways, Limitations

•
Takeaways:
◦
Presenting a new model (LQNet, CTRQNet) that overcomes the limitations of the static structure of existing quantum neural networks.
◦
Achieved up to 40% accuracy improvement over existing quantum neural networks (CIFAR-10 binary classification)
◦
Suggesting a potential contribution to understanding the black box of quantum machine learning.
•
Limitations:
◦
Additional evaluation of the proposed model's generalization performance and performance on various datasets is needed.
◦
Implementation and performance verification on actual quantum computers are required.
◦
Analysis of the model's complexity and computational cost is needed.
👍
Made with Slashpage