1. |
|
|
2. |
- Cichos, F., et al.
(author)
-
Machine learning for active matter
- 2020
-
In: Nature Machine Intelligence. - : Springer Science and Business Media LLC. - 2522-5839. ; 2:2, s. 94-103
-
Journal article (peer-reviewed)abstract
- The availability of large datasets has boosted the application of machine learning in many fields and is now starting to shape active-matter research as well. Machine learning techniques have already been successfully applied to active-matter data-for example, deep neural networks to analyse images and track objects, and recurrent nets and random forests to analyse time series. Yet machine learning can also help to disentangle the complexity of biological active matter, helping, for example, to establish a relation between genetic code and emergent bacterial behaviour, to find navigation strategies in complex environments, and to map physical cues to animal behaviours. In this Review, we highlight the current state of the art in the application of machine learning to active matter and discuss opportunities and challenges that are emerging. We also emphasize how active matter and machine learning can work together for mutual benefit. This Review surveys machine learning techniques that are currently developed for a range of research topics in biological and artificial active matter and also discusses challenges and exciting opportunities. This research direction promises to help disentangle the complexity of active matter and gain fundamental insights for instance in collective behaviour of systems at many length scales from colonies of bacteria to animal flocks.
|
|
3. |
- Erman, Eva, 1971-, et al.
(author)
-
The democratization of global AI governance and the role of tech companies
- 2024
-
In: Nature Machine Intelligence. - 2522-5839. ; 6, s. 246-248
-
Journal article (peer-reviewed)abstract
- Can non-state multinational tech companies counteract the potential democratic deficit in the emerging global governance of AI? We argue that although they may strengthen core values of democracy such as accountability and transparency, they currently lack the right kind of authority to democratize global AI governance.
|
|
4. |
- Gomariz, Alvaro, et al.
(author)
-
Modality attention and sampling enables deep learning with heterogeneous marker combinations in fluorescence microscopy
- 2021
-
In: Nature Machine Intelligence. - : Springer Nature. - 2522-5839. ; 3:9, s. 799-811
-
Journal article (peer-reviewed)abstract
- Fluorescence microscopy allows for a detailed inspection of cells, cellular networks and anatomical landmarks by staining with a variety of carefully selected markers visualized as colour channels. Quantitative characterization of structures in acquired images often relies on automatic image analysis methods. Despite the success of deep learning methods in other vision applications, their potential for fluorescence image analysis remains underexploited. One reason lies in the considerable workload required to train accurate models, which are normally specific for a given combination of markers and therefore applicable to a very restricted number of experimental settings. We herein propose ‘marker sampling and excite’—a neural network approach with a modality sampling strategy and a novel attention module that together enable (1) flexible training with heterogeneous datasets with combinations of markers and (2) successful utility of learned models on arbitrary subsets of markers prospectively. We show that our single neural network solution performs comparably to an upper bound scenario in which an ensemble of many networks is naively trained for each possible marker combination separately. We also demonstrate the feasibility of this framework in high-throughput biological analysis by revising a recent quantitative characterization of bone-marrow vasculature in three-dimensional confocal microscopy datasets and further confirm the validity of our approach on another substantially different dataset of microvessels in foetal liver tissues. Not only can our work substantially ameliorate the use of deep learning in fluorescence microscopy analysis, but it can also be utilized in other fields with incomplete data acquisitions and missing modalities.
|
|
5. |
- Guo, Jeff, et al.
(author)
-
Improving de novo molecular design with curriculum learning
- 2022
-
In: Nature Machine Intelligence. - : Springer Science and Business Media LLC. - 2522-5839. ; 4:6, s. 555-563
-
Journal article (peer-reviewed)abstract
- While reinforcement learning can be a powerful tool for complex design tasks such as molecular design, training can be slow when problems are either too hard or too easy, as little is learned in these cases. Jeff Guo and colleagues provide a curriculum learning extension to the REINVENT de novo molecular design framework that provides problems of increasing difficulty over epochs such that the training process is more efficient. Reinforcement learning is a powerful paradigm that has gained popularity across multiple domains. However, applying reinforcement learning may come at the cost of multiple interactions between the agent and the environment. This cost can be especially pronounced when the single feedback from the environment is slow or computationally expensive, causing extensive periods of non-productivity. Curriculum learning provides a suitable alternative by arranging a sequence of tasks of increasing complexity, with the aim of reducing the overall cost of learning. Here we demonstrate the application of curriculum learning for drug discovery. We implement curriculum learning in the de novo design platform REINVENT, and apply it to illustrative molecular design problems of different complexities. The results show both accelerated learning and a positive impact on the quality of the output when compared with standard policy-based reinforcement learning.
|
|
6. |
- Kc, G. B., et al.
(author)
-
A machine learning platform to estimate anti-SARS-CoV-2 activities
- 2021
-
In: Nature Machine Intelligence. - : Springer Science and Business Media LLC. - 2522-5839. ; 3, s. 527-535
-
Journal article (peer-reviewed)abstract
- Strategies for drug discovery and repositioning are urgently need with respect to COVID-19. Here we present REDIAL-2020, a suite of computational models for estimating small molecule activities in a range of SARS-CoV-2-related assays. Models were trained using publicly available, high-throughput screening data and by employing different descriptor types and various machine learning strategies. Here we describe the development and use of eleven models that span across the areas of viral entry, viral replication, live virus infectivity, in vitro infectivity and human cell toxicity. REDIAL-2020 is available as a web application through the DrugCentral web portal (http://drugcentral.org/Redial). The web application also provides similarity search results that display the most similar molecules to the query, as well as associated experimental data. REDIAL-2020 can serve as a rapid online tool for identifying active molecules for COVID-19 treatment.
|
|
7. |
- King, Ross, 1962, et al.
(author)
-
Cross-validation is safe to use
- 2021
-
In: Nature Machine Intelligence. - : Springer Science and Business Media LLC. - 2522-5839. ; 3:4, s. 276-276
-
Journal article (other academic/artistic)
|
|
8. |
- Niemiec, Emilia
(author)
-
A cautionary tale about the adoption of medical AI in Sweden
- 2023
-
In: Nature Machine Intelligence. - : Springer Science and Business Media LLC. - 2522-5839. ; 5:1, s. 5-7
-
Journal article (peer-reviewed)abstract
- A recent case of a flawed medical AI system that was backed by public funding provides an opportunity to discuss the impact of government policies and regulation in AI.
|
|
9. |
- Ollion, Etienne, 1980-, et al.
(author)
-
The dangers of using proprietary LLMs for research
- 2024
-
In: Nature Machine Intelligence. - : Nature Publishing Group. - 2522-5839. ; 6:1, s. 4-5
-
Journal article (peer-reviewed)abstract
- The release of ChatGPT at the end of 2022 thrust large language models (LLMs) into the limelight. By enabling its users to query the model directly in natural language, ChatGPT democratized access to these models — a welcome development. Since then, ChatGPT and similar tools such as Bard, Claude and Bing AI have shown their versatility and efficiency on a wide variety of tasks.
|
|
10. |
- Pek, Christian, et al.
(author)
-
Using online verification to prevent autonomous vehicles from causing accidents
- 2020
-
In: Nature Machine Intelligence. - : Nature Research. - 2522-5839. ; 2:9, s. 518-528
-
Journal article (peer-reviewed)abstract
- Ensuring that autonomous vehicles do not cause accidents remains a challenge. We present a formal verification technique for guaranteeing legal safety in arbitrary urban traffic situations. Legal safety means that autonomous vehicles never cause accidents although other traffic participants are allowed to perform any behaviour in accordance with traffic rules. Our technique serves as a safety layer for existing motion planning frameworks that provide intended trajectories for autonomous vehicles. We verify whether intended trajectories comply with legal safety and provide fallback solutions in safety-critical situations. The benefits of our verification technique are demonstrated in critical urban scenarios, which have been recorded in real traffic. The autonomous vehicle executed only safe trajectories, even when using an intended trajectory planner that was not aware of other traffic participants. Our results indicate that our online verification technique can drastically reduce the number of traffic accidents.
|
|