Featured
-
-
Perspective |
The data-driven future of high-energy-density physics
This Perspective discusses how high-energy-density physics could tap the potential of AI-inspired algorithms for extracting relevant information and how data-driven automatic control routines may be used for optimizing high-repetition-rate experiments.
- Peter W. Hatfield
- , Jim A. Gaffney
- & Ben Williams
-
Matters Arising |
Reply to: Concerns about phytoplankton bloom trends in global lakes
- Jeff C. Ho
- , Anna M. Michalak
- & Nima Pahlevan
-
Article |
Bayesian reaction optimization as a tool for chemical synthesis
Bayesian optimization is applied in chemical synthesis towards the optimization of various organic reactions and is found to outperform scientists in both average optimization efficiency and consistency.
- Benjamin J. Shields
- , Jason Stevens
- & Abigail G. Doyle
-
Article |
Widespread global increase in intense lake phytoplankton blooms since the 1980s
Analyses show that the peak intensity of summertime phytoplankton blooms has increased in 71 large lakes globally over the past three decades, revealing a worldwide exacerbation of bloom conditions.
- Jeff C. Ho
- , Anna M. Michalak
- & Nima Pahlevan
-
Letter |
Anthropogenic biases in chemical reaction data hinder exploratory inorganic synthesis
Human scientists make unrepresentative chemical reagent and reaction condition choices, and machine-learning algorithms trained on human-selected experiments are less capable of successfully predicting reaction outcomes than those trained on randomly generated experiments.
- Xiwen Jia
- , Allyson Lynch
- & Joshua Schrier
-
Letter |
Unsupervised word embeddings capture latent knowledge from materials science literature
Natural language processing algorithms applied to three million materials science abstracts uncover relationships between words, material compositions and properties, and predict potential new thermoelectric materials.
- Vahe Tshitoyan
- , John Dagdelen
- & Anubhav Jain
-
Letter |
Predicting disruptive instabilities in controlled fusion plasmas through deep learning
Using data from plasma-based tokamak nuclear reactors in the US and Europe, a machine-learning approach based on deep neural networks is taught to forecast disruptions, even those in machines on which the algorithm was not trained.
- Julian Kates-Harbeck
- , Alexey Svyatkovskiy
- & William Tang
-
Letter |
Compensatory water effects link yearly global land CO2 sink changes to temperature
A study of how temperature and water availability fluctuations affect the carbon balance of land ecosystems reveals different controls on local and global scales, implying that spatial climate covariation drives the global carbon cycle response.
- Martin Jung
- , Markus Reichstein
- & Ning Zeng
-
Letter |
Reconstructing state mixtures from diffraction measurements
An imaging technique has been developed to characterize state mixtures caused by partial coherence and fluctuations in dynamical systems.
- Pierre Thibault
- & Andreas Menzel
-
-
-
News |
Physicists count on updated constants
The latest revision of fundamental quantities bodes well for the proposed overhaul of the international system of units.
- Eugenie Samuel Reich
-
News |
Researchers launch hunt for endangered data
Global effort will catalogue information languishing in drawers and basements.
- Linda Nordling
-
Correspondence |
New data system to galvanize Brazil's conservation efforts
- Ana C. M. Malhado
- & Richard J. Ladle
Identifying general reaction conditions by bandit optimization