Artificial and Augmented Intelligence for Automated Scientific Discovery

用于自动化科学发现的人工智能和增强智能

• 批准号: EP/S000356/1
• 负责人: Jeremy Frey
• 金额: $ 129.24万
• 依托单位: University of Southampton
• 依托单位国家: 英国
• 项目类别: Research Grant
• 财政年份: 2018
• 资助国家: 英国
• 起止时间: 2018 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=EP%2FS000356%2F1
• 关键词: Artificial; Augmented; Intelligence; Automated; Scientific

AI is a widely used term that conjurers up many of the computers from science fiction. Its stands for a whole collection of ideas, algorithms, computational models and knowledge systems. Recent success of particular types of machine learning (e.g. deep neutral nets) have again excited the interest of the scientific community in delivering insight into the complexity of the real world. This type of approach compliments the knowledge engineering systems that have previously been used, however they require massive amounts of data to be trained. Taking the chemical and materials sciences as exemplar areas we can see that the traditional approaches to scientific discovery work with relatively small amounts of often uncertain data which is distilled by human insight to yield predictions and testable theories which may evolve as new data becomes available. In these areas of science more data is becoming available and the impact of 'larger data' parallels the reality that almost all science now depends on computational assistance. Never-the-less the quantity of quality data needed to train the new AI systems is simply not directly available even with recent advances in automation. As a basis for the network we propose to use 'amplification by simulation' as a key element of the cycle of automated experiments, simulation, AI learning, prediction, comparison, design, further experiments, to create the environment in which leading AI developments can be applied to the chemical and materials discovery.

人工智能是一个广泛使用的术语，它使许多科幻小说中的计算机变戏法。它代表了思想、算法、计算模型和知识系统的整体集合。最近，特定类型的机器学习（例如深度中性网络）的成功再次激发了科学界对洞察真实的世界复杂性的兴趣。这种方法补充了以前使用的知识工程系统，但它们需要大量的数据进行训练。以化学和材料科学为例，我们可以看到，传统的科学发现方法只处理相对少量的、通常不确定的数据，这些数据是通过人类的洞察力提炼出来的，可以产生预测和可检验的理论，这些理论可能会随着新数据的出现而发展。在这些科学领域，越来越多的数据变得可用，“更大数据”的影响与几乎所有科学现在都依赖于计算辅助的现实平行。尽管如此，即使在自动化方面取得了最新进展，训练新人工智能系统所需的高质量数据量也无法直接获得。作为网络的基础，我们建议使用“模拟放大”作为自动化实验，模拟，人工智能学习，预测，比较，设计，进一步实验循环的关键要素，以创建领先的人工智能发展可以应用于化学和材料发现的环境。

项目成果

Group 8: Challenge: Event detection in nanopore data

第 8 组：挑战：纳米孔数据中的事件检测

• DOI: 10.5258/soton/ai3sd0249
• 发表时间: 2022
• 作者: Adewole W

Group 5: Challenge: Task 2 - Event Detection in Nanopore Data

第 5 组：挑战：任务 2 - 纳米孔数据中的事件检测

• 发表时间: 2022
• 作者: Bachs Herra, A

Group 14: Challenge: Task 3 - Defect Detection in Graphene Sheets

第 14 组：挑战：任务 3 - 石墨烯片中的缺陷检测

• 发表时间: 2022
• 作者: Bao P

AI3SD Project: Artificial intelligence for reconstruction and super-resolution of chemical tomography

AI3SD项目：用于化学断层扫描重建和超分辨率的人工智能

• 发表时间: 2021
• 作者: Butler KT

AI3SD Intern Project: A deep neural network for structural relaxation of metal-organic interfaces

AI3SD 实习生项目：用于金属有机界面结构松弛的深度神经网络

• 发表时间: 2021
• 作者: Barret R