Scholarlink

<p>This paper presents a framework for converting a classical recurrent neural network (RNN) into a quantum neural network (QNN), demonstrating the potential advantages of quantum mechanics in neural computation. By mapping classical neurons, activation functions, and recurrent dynamics into their quantum analogs, we propose a quantum Hamiltonian that governs the system’s time evolution, converting the features of the classical system in a quantum mechanical context. We analyze the effects of key parameters, particularly the interaction coefficient <InlineEquation ID="IEq1"> <InlineMediaObject> <ImageObject Color="BlackWhite" FileRef="11128_2025_4806_Article_IEq1.gif" Format="GIF" Height="17" Rendition="HTML" Resolution="72" Type="Linedraw" Width="15" /> </InlineMediaObject> <EquationSource Format="TEX">\(\beta \)</EquationSource> <EquationSource Format="MATHML"><math> <mi>β</mi> </math></EquationSource> </InlineEquation>, on the system’s dynamical behavior, showing oscillatory patterns in the expectation values of Pauli matrices. Furthermore, we incorporate quantum noise and explore the quantum analog of classical inputs. Our findings show State purity decays slowly, and the energy expectation exhibits damped oscillations—both confirming a rich balance of unitary processing and controlled decoherence, showing interesting departure from classical recurrent networks.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Mapping additive recurrent neural networks to quantum neural networks

  • 期刊论文

  • 发表日期：

    2025年6月17日

查看全文
  • Rakesh Sengupta

摘要

This paper presents a framework for converting a classical recurrent neural network (RNN) into a quantum neural network (QNN), demonstrating the potential advantages of quantum mechanics in neural computation. By mapping classical neurons, activation functions, and recurrent dynamics into their quantum analogs, we propose a quantum Hamiltonian that governs the system’s time evolution, converting the features of the classical system in a quantum mechanical context. We analyze the effects of key parameters, particularly the interaction coefficient \(\beta \) β , on the system’s dynamical behavior, showing oscillatory patterns in the expectation values of Pauli matrices. Furthermore, we incorporate quantum noise and explore the quantum analog of classical inputs. Our findings show State purity decays slowly, and the energy expectation exhibits damped oscillations—both confirming a rich balance of unitary processing and controlled decoherence, showing interesting departure from classical recurrent networks.