Found a total of 8 related content
打开AI黑匣子,「三段式」AI用于化学研究,优化分子同时产生新化学知识,登Nature
Article Introduction:编辑|KXAI工具的强大功能AI工具的功能令人难以置信。然而,如果你想要深入了解其运作方式,通常会毫无头绪。AI常常被视为「黑匣子」。AI在化学中的应用对于化学来说,AI可以帮助我们优化分子,但它无法告诉我们为何这是最佳的——重要的是什么特性、结构和功能?伊利诺伊大学厄巴纳-香槟分校的研究创新近日,伊利诺伊大学厄巴纳-香槟分校(UIUC)的一个跨学科研究团队打开了黑匣子。研究人员通过将AI与自动化学合成和实验验证相结合,找到了AI所依赖的化学原理,从而改进用于收集太阳能的分子。该研究发现了比现有
2024-09-06
comment 0
410
Learn performance optimization and code analysis in JavaScript
Article Introduction:Learning performance optimization and code analysis in JavaScript requires concrete code examples. JavaScript is a widely used programming language in web development. As application complexity increases and users' performance requirements continue to increase, how to optimize the performance of JavaScript code has become an important challenge faced by developers. This article will introduce several common JavaScript performance optimization techniques and provide specific code examples for readers' reference. First, optimizing the scope of variables is to increase J
2023-11-03
comment 0
916
「两全其美」,从头设计分子,深度学习架构S4用于化学语言建模
Article Introduction:编辑|KX生成式深度学习正在重塑药物设计。化学语言模型(CLM)以分子串的形式生成分子,对这一过程尤为重要。近日,来自荷兰埃因霍芬理工大学(EindhovenUniversityofTechnology)的研究人员将一种最新的深度学习架构(S4)引入到从头药物设计中。结构化状态空间序列(StructuredStateSpaceSequence,S4)模型在学习序列的全局属性方面表现卓越,那么S4能否推进从头设计的化学语言建模?为了给出答案,研究人员系统地在一系列药物发现任务上对S4与最先进的CLM进行了基
2024-08-06
comment
495
Data processing and formatting experience sharing in JavaScript development
Article Introduction:JavaScript is a scripting language widely used in web development. It is dynamic and flexible. In JavaScript development, data processing and formatting are very important parts. This article will share some experiences in data processing and formatting in JavaScript development. 1. Data processing In JavaScript development, various types of data need to be processed, including strings, numbers, arrays, and objects. Here are some common data processing techniques: String Processing: Ja
2023-11-02
comment 0
347
Peking University & Wangshi Intelligence proposes a new model: bridging the gap between chemical reaction pre-training and conditional molecule generation!
Article Introduction:Chemical reactions are the basis of drug design and organic chemistry research. There is a growing need among the research community for a large-scale deep learning framework that can effectively capture the fundamental rules of chemical reactions. Recently, a research team from Peking University and Wangshi Intelligence proposed a new method to bridge the gap between reaction-based molecular pre-training and generation tasks. Inspired by the mechanisms of organic chemistry, researchers developed a new pre-training framework that enables it to incorporate inductive bias into models. This proposed framework achieves state-of-the-art results when performing challenging downstream tasks. By leveraging knowledge of chemistry, the framework overcomes the limitations of current molecular generation models that rely on a small number of reaction templates. In extensive experiments, the model generated high-quality synthesizable drug-like structures. Overall, the model
2023-12-14
comment 0
262
厦大团队材料预测迁移学习范式登Nature子刊,发现高性能催化剂
Article Introduction:编辑|KX传统的材料发现依赖反复试验或偶然发现,效率低下且成本高昂。AI在发现新型催化剂方面潜力巨大。然而,受到算法的选择,以及数据质量和数量的影响。在此,来自厦门大学、深圳大学、武汉大学、南京航空航天大学和英国利物浦大学的研究团队开发了一种迁移学习范式,结合了预训练模型、集成学习和主动学习,能够预测未被发现的钙钛矿氧化物,并增强该反应的通用性。通过筛选16,050种成分,鉴定和合成了36种新的钙钛矿氧化物,其中包括13种纯钙钛矿结构。Pr0.1Sr0.9Co0.5Fe0.5O3(PSCF)和Pr0.1S
2024-07-31
comment
947
Distributed training and model parallelization technology and applications in deep learning using Java
Article Introduction:With the continuous development of computing technology and the continuous maturity of deep learning algorithms, deep learning has gradually become a popular technology in the field of machine learning. When performing deep learning training, if you only use a single computer for training, it may take a very long time, and the training data requires a relatively large amount of memory. In order to perform deep learning training efficiently, we need to make full use of computing resources, which requires the application of distributed training technology and model parallelization technology. This article will discuss the methods and applications of implementing these technologies using Java. Distributed training and
2023-06-18
comment 0
666