<p>In modern neural network research, quantum neural network (QNN) methodologies have been widely adopted due to their inherent quantum advantages. QNN architectures can be partitioned, with each segment distributed across multiple computing machines to preserve privacy-an approach referred to as quantum split learning. Although several novel QNN software testing techniques have been introduced, no methodology currently exists for testing quantum distributed learning software such as quantum split learning. To address this issue, this paper proposes a new software testing method for quantum distributed/split learning, named SQUAD. The proposed SQUAD automatically generates dummy code to complete QNN architectures across separate machines and subsequently performs software testing on all machines. Additionally, a graphical user interface (GUI) for SQUAD is implemented to demonstrate its novelty and feasibility.</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

SQUAD: software testing for quantum distributed learning software

  • Soohyun Park,
  • Jae Hyun Cho,
  • Hyun Jun Yook,
  • Ga San Jhun,
  • Youn Kyu Lee,
  • Joongheon Kim

摘要

In modern neural network research, quantum neural network (QNN) methodologies have been widely adopted due to their inherent quantum advantages. QNN architectures can be partitioned, with each segment distributed across multiple computing machines to preserve privacy-an approach referred to as quantum split learning. Although several novel QNN software testing techniques have been introduced, no methodology currently exists for testing quantum distributed learning software such as quantum split learning. To address this issue, this paper proposes a new software testing method for quantum distributed/split learning, named SQUAD. The proposed SQUAD automatically generates dummy code to complete QNN architectures across separate machines and subsequently performs software testing on all machines. Additionally, a graphical user interface (GUI) for SQUAD is implemented to demonstrate its novelty and feasibility.