Planning: aiNsect: A Digital Insect Brain

规划：aiNsect：数字昆虫大脑

• 批准号: 2332218
• 负责人: Chase Stratton
• 金额: $ 9.91万
• 依托单位: Delaware State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-10-01 至 2024-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2332218&HistoricalAwards=false
• 关键词: Planning; aiNsect; Digital; Insect; Brain

Insects have relatively simple brains, and chemicals are one of the strongest external stimuli that affect their behaviors. For insects, external molecules largely dictate behaviors such as attraction, repellence, change of direction or action, and cessation of activity. Predicting insect behavior in response to chemicals is vital to insect and neurobehavioral sciences, including understanding production of organic and conventional insecticides, finding natural repellents for biting insects, or isolating pheromones to attract beneficial insects. To address these questions, the project long-term goal is to develop an artificial insect brain called aiNsect by using machine learning techniques and the neuro-physiological map of fruit fly brains. The planning grant activities will be an important first step by supporting feasibility studies and establishing collaborations. By training aiNsect to interpret how insects respond to chemicals, it will uncover the patterns and rules that guide those behaviors. This interdisciplinary research, combined with ever-expanding advancements in AI, will pave the way for learning programs that comprehend even subtle shifts in behavior and the neurons involved. The project's primary objectives include: (A) settling on an assay design that captures even the smallest movements of all test subjects; (B) fostering collaboration across our network of Historically Black Colleges and Universities, large research institution, non-profit and commercial organizations and government agencies, R01s, non-profits, government agencies, and commercial organizations; and (C) proving the feasibility of the AI we propose with published code. This work will provide a fundamental base to generate a research plan leading to a full submission to the HBCU-Excellence in Research Program.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

昆虫的大脑相对简单，化学物质是影响其行为的最强外部刺激之一。对于昆虫来说，外部分子在很大程度上决定了诸如吸引、排斥、改变方向或动作以及停止活动等行为。 预测昆虫对化学物质的反应行为对昆虫和神经行为科学至关重要，包括了解有机和常规杀虫剂的生产，寻找叮咬昆虫的天然驱虫剂，或分离信息素以吸引有益昆虫。为了解决这些问题，该项目的长期目标是通过使用机器学习技术和果蝇大脑的神经生理学图谱来开发一种名为aiNsect的人工昆虫大脑。规划赠款活动将是重要的第一步，支持可行性研究和建立合作。通过训练人工智能来解释昆虫对化学物质的反应，它将揭示指导这些行为的模式和规则。这种跨学科的研究，再加上人工智能不断扩大的进步，将为学习程序铺平道路，这些程序甚至可以理解行为和相关神经元的微妙变化。该项目的主要目标包括：（A）确定一种能够捕捉所有测试对象的最小运动的检测设计;（B）促进我们的历史黑人学院和大学、大型研究机构、非营利和商业组织以及政府机构、R 01、非营利组织、政府机构和商业组织网络之间的合作;以及（C）用公开的代码证明我们提出的人工智能的可行性。这项工作将提供一个基本的基础，以产生一个研究计划，导致一个完整的提交给HBCU-卓越的研究计划。这个奖项反映了NSF的法定使命，并已被认为是值得通过评估使用基金会的知识价值和更广泛的影响审查标准的支持。