人工智能在生物多样性保护中可信度研究博士研究员
PhD Studentship: Can we Trust AI to Help Save Biodiversity?
原帖优先:申请材料、截止时间与资格以原帖和学校官方说明为准。
AI 中文速览
- 研究内容
- 本项目旨在研究AI自动化监测生态证据的可靠性及改进方法,结合大型生物多样性数据集、模拟和统计建模,探讨AI面临挑战的物种与环境条件,量化检测误差对种群占据、丰度和趋势估计的影响,并测试生态背景或互补传感器是否能提升可靠性。
- 申请条件
- 原文未说明具体学位门槛,但通常要求相关领域的学术背景及数据分析与生态学/计算机科学相关能力。
- 待遇
- 对于符合条件的学者,该奖学金将涵盖英国本土学费,并提供每年免税的生活补助。
- 申请方式
- 原文未说明具体的申请方式、截止日期及提交地址。
由 gemini-2.5-flash-lite 生成,博士岗判定置信度 100%。
岗位信息
- 截止
- (Europe/London) 剩 45 天
- 学科
- 生命科学
- 合同类型
- 雇佣合同
- 原帖发布
- 本站收录
- 内容更新
- 原帖导师候选(待核验)
- Benno Simmons, Sareh Rowlands, Chris Clements
- 来源
- jobs.ac.uk(英国博士项目及学术招聘) · 最近核对 2026-11-22
- 详情核验
判定依据(原文摘录)
- 这是一个博士生奖学金项目
PhD Studentship: Can we Trust AI to Help Save Biodiversity?
- 奖学金涵盖学费与生活补助
For eligible students the studentship will cover home tuition fees plus an annual tax-free stipend
原文
AI is transforming how we monitor biodiversity, but could errors in what machines detect lead us to the wrong conservation decisions? This project will establish when automated monitoring provides reliable ecological evidence and how to improve it. Combining large biodiversity datasets with simulation and statistical modelling, the student will investigate which species and environmental conditions challenge AI, quantify how detection errors affect estimates of occupancy, abundance and population trends, and test whether ecological context or complementary sensors improve reliability.
The student will examine how traits, taxonomy, habitat, season, environmental conditions and training-data availability shape AI performance, then track how false positives, false negatives and uncertainty influence ecological conclusions. They will also explore ways to make automated monitoring more trustworthy, for example by using ecological information such as location, habitat and season to improve uncertainty estimates, or by combining cameras, acoustics, eDNA and remote sensing. The project has a clear core structure, but the student will be strongly encouraged to develop their own ideas, choose focal taxa and datasets, and shape the final direction of the research.
Training will cover reproducible data science, AI and computer vision, hierarchical ecological modelling, uncertainty quantification, high-performance computing, scientific writing and communication. Benno Simmons (Exeter, Ecology), Sareh Rowlands (Exeter, Computer Science) and Chris Clements (Bristol, Ecology) will provide hands-on training through coding, joint analysis, research meetings and paper writing. The student will also gain experience communicating technical results to conservation organisations, local environmental partnerships and policy audiences, learning how to turn methodological advances into evidence that can be used in practice.
The Exeter–Bristol collaboration brings together complementary expertise in ecology and computer science. Joint supervision, regular interaction and research visits to Bristol will give the student access to both research communities, broaden their networks and support them in developing an independent research profile. For eligible students the studentship will cover home tuition fees plus an annual tax-free stipend