Nature Machine Intelligence最新文献

{"title":"Back to basics to open the black box","authors":"Diego Marcondes, Adilson Simonis, Junior Barrera","doi":"10.1038/s42256-024-00842-6","DOIUrl":"10.1038/s42256-024-00842-6","url":null,"abstract":"Most research efforts in machine learning focus on performance and are detached from an explanation of the behaviour of the model. We call for going back to basics of machine learning methods, with more focus on the development of a basic understanding grounded in statistical theory.","PeriodicalId":48533,"journal":{"name":"Nature Machine Intelligence","volume":null,"pages":null},"PeriodicalIF":23.8,"publicationDate":"2024-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140953389","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Does it matter if empathic AI has no empathy?","authors":"Garriy Shteynberg, Jodi Halpern, Amir Sadovnik, Jon Garthoff, Anat Perry, Jessica Hay, Carlos Montemayor, Michael A. Olson, Tim L. Hulsey, Abrol Fairweather","doi":"10.1038/s42256-024-00841-7","DOIUrl":"10.1038/s42256-024-00841-7","url":null,"abstract":"","PeriodicalId":48533,"journal":{"name":"Nature Machine Intelligence","volume":null,"pages":null},"PeriodicalIF":23.8,"publicationDate":"2024-05-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140925029","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-purpose RNA language modelling with motif-aware pretraining and type-guided fine-tuning","authors":"Ning Wang, Jiang Bian, Yuchen Li, Xuhong Li, Shahid Mumtaz, Linghe Kong, Haoyi Xiong","doi":"10.1038/s42256-024-00836-4","DOIUrl":"10.1038/s42256-024-00836-4","url":null,"abstract":"Pretrained language models have shown promise in analysing nucleotide sequences, yet a versatile model excelling across diverse tasks with a single pretrained weight set remains elusive. Here we introduce RNAErnie, an RNA-focused pretrained model built upon the transformer architecture, employing two simple yet effective strategies. First, RNAErnie enhances pretraining by incorporating RNA motifs as biological priors and introducing motif-level random masking in addition to masked language modelling at base/subsequence levels. It also tokenizes RNA types (for example, miRNA, lnRNA) as stop words, appending them to sequences during pretraining. Second, subject to out-of-distribution tasks with RNA sequences not seen during the pretraining phase, RNAErnie proposes a type-guided fine-tuning strategy that first predicts possible RNA types using an RNA sequence and then appends the predicted type to the tail of sequence to refine feature embedding in a post hoc way. Our extensive evaluation across seven datasets and five tasks demonstrates the superiority of RNAErnie in both supervised and unsupervised learning. It surpasses baselines with up to 1.8% higher accuracy in classification, 2.2% greater accuracy in interaction prediction and 3.3% improved F1 score in structure prediction, showcasing its robustness and adaptability with a unified pretrained foundation. Despite the existence of various pretrained language models for nucleotide sequence analysis, achieving good performance on a broad range of downstream tasks using a single model is challenging. Wang and colleagues develop a pretrained language model specifically optimized for RNA sequence analysis and show that it can outperform state-of-the-art methods in a diverse set of downstream tasks.","PeriodicalId":48533,"journal":{"name":"Nature Machine Intelligence","volume":null,"pages":null},"PeriodicalIF":23.8,"publicationDate":"2024-05-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.nature.com/articles/s42256-024-00836-4.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140919490","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Diving into deep learning","authors":"Ge Wang","doi":"10.1038/s42256-024-00840-8","DOIUrl":"10.1038/s42256-024-00840-8","url":null,"abstract":"","PeriodicalId":48533,"journal":{"name":"Nature Machine Intelligence","volume":null,"pages":null},"PeriodicalIF":23.8,"publicationDate":"2024-05-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140903023","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Predicting equilibrium distributions for molecular systems with deep learning","authors":"Shuxin Zheng, Jiyan He, Chang Liu, Yu Shi, Ziheng Lu, Weitao Feng, Fusong Ju, Jiaxi Wang, Jianwei Zhu, Yaosen Min, He Zhang, Shidi Tang, Hongxia Hao, Peiran Jin, Chi Chen, Frank Noé, Haiguang Liu, Tie-Yan Liu","doi":"10.1038/s42256-024-00837-3","DOIUrl":"10.1038/s42256-024-00837-3","url":null,"abstract":"Advances in deep learning have greatly improved structure prediction of molecules. However, many macroscopic observations that are important for real-world applications are not functions of a single molecular structure but rather determined from the equilibrium distribution of structures. Conventional methods for obtaining these distributions, such as molecular dynamics simulation, are computationally expensive and often intractable. Here we introduce a deep learning framework, called Distributional Graphormer (DiG), in an attempt to predict the equilibrium distribution of molecular systems. Inspired by the annealing process in thermodynamics, DiG uses deep neural networks to transform a simple distribution towards the equilibrium distribution, conditioned on a descriptor of a molecular system such as a chemical graph or a protein sequence. This framework enables the efficient generation of diverse conformations and provides estimations of state densities, orders of magnitude faster than conventional methods. We demonstrate applications of DiG on several molecular tasks, including protein conformation sampling, ligand structure sampling, catalyst–adsorbate sampling and property-guided structure generation. DiG presents a substantial advancement in methodology for statistically understanding molecular systems, opening up new research opportunities in the molecular sciences. Methods for predicting molecular structure predictions have so far focused on only the most probable conformation, but molecular structures are dynamic and can change when performing their biological functions, for example. Zheng et al. use a graph transformer approach to learn the equilibrium distribution of molecular systems and show that this can be helpful for a number of downstream tasks, including protein structure prediction, ligand docking and molecular design.","PeriodicalId":48533,"journal":{"name":"Nature Machine Intelligence","volume":null,"pages":null},"PeriodicalIF":23.8,"publicationDate":"2024-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.nature.com/articles/s42256-024-00837-3.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140890512","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Augmenting large language models with chemistry tools","authors":"Andres M. Bran, Sam Cox, Oliver Schilter, Carlo Baldassari, Andrew D. White, Philippe Schwaller","doi":"10.1038/s42256-024-00832-8","DOIUrl":"10.1038/s42256-024-00832-8","url":null,"abstract":"Large language models (LLMs) have shown strong performance in tasks across domains but struggle with chemistry-related problems. These models also lack access to external knowledge sources, limiting their usefulness in scientific applications. We introduce ChemCrow, an LLM chemistry agent designed to accomplish tasks across organic synthesis, drug discovery and materials design. By integrating 18 expert-designed tools and using GPT-4 as the LLM, ChemCrow augments the LLM performance in chemistry, and new capabilities emerge. Our agent autonomously planned and executed the syntheses of an insect repellent and three organocatalysts and guided the discovery of a novel chromophore. Our evaluation, including both LLM and expert assessments, demonstrates ChemCrow’s effectiveness in automating a diverse set of chemical tasks. Our work not only aids expert chemists and lowers barriers for non-experts but also fosters scientific advancement by bridging the gap between experimental and computational chemistry. Large language models can be queried to perform chain-of-thought reasoning on text descriptions of data or computational tools, which can enable flexible and autonomous workflows. Bran et al. developed ChemCrow, a GPT-4-based agent that has access to computational chemistry tools and a robotic chemistry platform, which can autonomously solve tasks for designing or synthesizing chemicals such as drugs or materials.","PeriodicalId":48533,"journal":{"name":"Nature Machine Intelligence","volume":null,"pages":null},"PeriodicalIF":23.8,"publicationDate":"2024-05-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.nature.com/articles/s42256-024-00832-8.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140890495","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Maximum diffusion reinforcement learning","authors":"Thomas A. Berrueta, Allison Pinosky, Todd D. Murphey","doi":"10.1038/s42256-024-00829-3","DOIUrl":"10.1038/s42256-024-00829-3","url":null,"abstract":"Robots and animals both experience the world through their bodies and senses. Their embodiment constrains their experiences, ensuring that they unfold continuously in space and time. As a result, the experiences of embodied agents are intrinsically correlated. Correlations create fundamental challenges for machine learning, as most techniques rely on the assumption that data are independent and identically distributed. In reinforcement learning, where data are directly collected from an agent’s sequential experiences, violations of this assumption are often unavoidable. Here we derive a method that overcomes this issue by exploiting the statistical mechanics of ergodic processes, which we term maximum diffusion reinforcement learning. By decorrelating agent experiences, our approach provably enables single-shot learning in continuous deployments over the course of individual task attempts. Moreover, we prove our approach generalizes well-known maximum entropy techniques and robustly exceeds state-of-the-art performance across popular benchmarks. Our results at the nexus of physics, learning and control form a foundation for transparent and reliable decision-making in embodied reinforcement learning agents. The central assumption in machine learning that data are independent and identically distributed does not hold in many reinforcement learning settings, as experiences of reinforcement learning agents are sequential and intrinsically correlated in time. Berrueta and colleagues use the mathematical theory of ergodic processes to develop a reinforcement framework that can decorrelate agent experiences and is capable of learning in single-shot deployments.","PeriodicalId":48533,"journal":{"name":"Nature Machine Intelligence","volume":null,"pages":null},"PeriodicalIF":23.8,"publicationDate":"2024-05-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140819404","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The rewards of reusable machine learning code","authors":"","doi":"10.1038/s42256-024-00835-5","DOIUrl":"10.1038/s42256-024-00835-5","url":null,"abstract":"Research papers can make a long-lasting impact when the code and software tools supporting the findings are made readily available and can be reused and built on. Our reusability reports explore and highlight examples of good code sharing practices.","PeriodicalId":48533,"journal":{"name":"Nature Machine Intelligence","volume":null,"pages":null},"PeriodicalIF":23.8,"publicationDate":"2024-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.nature.com/articles/s42256-024-00835-5.pdf","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140642056","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The benefits, risks and bounds of personalizing the alignment of large language models to individuals","authors":"Hannah Rose Kirk, Bertie Vidgen, Paul Röttger, Scott A. Hale","doi":"10.1038/s42256-024-00820-y","DOIUrl":"10.1038/s42256-024-00820-y","url":null,"abstract":"Large language models (LLMs) undergo ‘alignment’ so that they better reflect human values or preferences, and are safer or more useful. However, alignment is intrinsically difficult because the hundreds of millions of people who now interact with LLMs have different preferences for language and conversational norms, operate under disparate value systems and hold diverse political beliefs. Typically, few developers or researchers dictate alignment norms, risking the exclusion or under-representation of various groups. Personalization is a new frontier in LLM development, whereby models are tailored to individuals. In principle, this could minimize cultural hegemony, enhance usefulness and broaden access. However, unbounded personalization poses risks such as large-scale profiling, privacy infringement, bias reinforcement and exploitation of the vulnerable. Defining the bounds of responsible and socially acceptable personalization is a non-trivial task beset with normative challenges. This article explores ‘personalized alignment’, whereby LLMs adapt to user-specific data, and highlights recent shifts in the LLM ecosystem towards a greater degree of personalization. Our main contribution explores the potential impact of personalized LLMs via a taxonomy of risks and benefits for individuals and society at large. We lastly discuss a key open question: what are appropriate bounds of personalization and who decides? Answering this normative question enables users to benefit from personalized alignment while safeguarding against harmful impacts for individuals and society. Tailoring the alignment of large language models (LLMs) to individuals is a new frontier in generative AI, but unbounded personalization can bring potential harm, such as large-scale profiling, privacy infringement and bias reinforcement. Kirk et al. develop a taxonomy for risks and benefits of personalized LLMs and discuss the need for normative decisions on what are acceptable bounds of personalization.","PeriodicalId":48533,"journal":{"name":"Nature Machine Intelligence","volume":null,"pages":null},"PeriodicalIF":23.8,"publicationDate":"2024-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140636129","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Dangers of speech technology for workplace diversity","authors":"Mike Horia Mihail Teodorescu, Mingang K. Geiger, Lily Morse","doi":"10.1038/s42256-024-00827-5","DOIUrl":"10.1038/s42256-024-00827-5","url":null,"abstract":"Speech technology offers many applications to enhance employee productivity and efficiency. Yet new dangers arise for marginalized groups, potentially jeopardizing organizational efforts to promote workplace diversity. Our analysis delves into three critical risks of speech technology and offers guidance for mitigating these risks responsibly.","PeriodicalId":48533,"journal":{"name":"Nature Machine Intelligence","volume":null,"pages":null},"PeriodicalIF":23.8,"publicationDate":"2024-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"140632263","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}