in

Gran Turismo champion, reimagined urine — the week in infographics

[adace-ad id="91168"]

  • NEWS
  • 15 February 2022

Gran Turismo champion, reimagined urine — the week in infographics

  • Twitter

  • Facebook

  • Email

Artificial intelligence overtakes human gamers

This graphic shows one way in which an artificial intelligence (AI) is able to win against the best human players of the video game Gran Turismo. In a paper in Nature, a team of researchers introduce GT Sophy, which learns through a neural-network model.

GT Sophy stands out for its performance against human drivers in a head-to-head competition. Far from using a lap-time advantage to outlast opponents, GT Sophy simply outraces them. Through the training process, GT Sophy learnt to take different lines through the corners in response to different conditions. Our graphic shows how, in one case, two human drivers attempted to block the preferred path of two GT Sophy cars, yet the AI succeeded in finding two trajectories that overcame this block and allowed its cars to overtake. You can read more about what it takes to win at racing (both real and simulated) in this News & Views article.

The march of methane

Levels of methane, a potent greenhouse gas, have been growing for decades — but they began a rapid and mysterious uptick around 2007. Last year, methane concentrations in the atmosphere raced past 1,900 parts per billion, nearly triple pre-industrial levels, according to data released in January by the US National Oceanic and Atmospheric Administration.

Where is it coming from? Potential explanations range from the expanding exploitation of oil and natural gas, and rising emissions from landfill, to growing livestock herds and increasing activity by microbes in wetlands. The spike has caused many researchers to worry that global warming is creating a feedback mechanism that will cause ever more methane to be released, making it even harder to rein in rising global temperatures.

Source: NOAA

Urine, reimagined

Our final graphic this week illustrates some of the many ways in which human urine could be recycled into useful products. Scientists say that urine diversion would have huge environmental and public-health benefits if deployed on a large scale. That’s in part because urine is rich in nutrients that could help to fertilize crops or feed into industrial processes; furthermore, not flushing urine down the drain could save vast amounts of water.

But urine diversion and reuse would require “drastic reimagining of how we do human sanitation”, as a Feature reports. It would involve wide-scale use of special urine-diverting toilets, and even processing devices in your building’s basement.

doi: https://doi.org/10.1038/d41586-022-00458-z

  • Neural networks overtake humans in Gran Turismo racing game

  • Scientists raise alarm over ‘dangerously fast’ growth in atmospheric methane

  • The urine revolution: how recycling pee could help to save the world

Subjects

  • Machine learning
  • Water resources
  • Climate change

Latest on:

Neural networks overtake humans in Gran Turismo racing game

News & Views

Early prediction of preeclampsia in pregnancy with cell-free RNA

Article

DeepMind AI tackles one of chemistry’s most valuable techniques

News

The urine revolution: how recycling pee could help to save the world

News Feature

‘Sky river’ brought Iran deadly floods — but also welcome water

Research Highlight

Brazil is in water crisis — it needs a drought plan

Comment

Scientists raise alarm over ‘dangerously fast’ growth in atmospheric methane

News

Climate pledges from top companies crumble under scrutiny

News

Emissions cuts take political and social innovation too

Correspondence

Jobs

  • Flow cytometrist – Sorter Operator (m/f/div)

    Max Planck Institute for Biology of Ageing (MPIAGE)

    Köln, Germany

  • Physicist, Engineer Electrical Engineering and/or photonics or similar (f/m/x) – Doctoral Fellowship – Optical technologies for space navigation

    German Aerospace Center (DLR)

    Oberpfaffenhofen, Germany

  • Doctoral Researcher (f/m/d) 75 % for the project C2 Seismic imaging by full waveform inversion

    Karlsruhe Institute of Technology (KIT)

    Karlsruhe, Germany

  • Research associate / PhD candidate (f/m/d) Embedded Electronic Systems

    Karlsruhe Institute of Technology (KIT)

    Karlsruhe, Germany


Source: Resources - nature.com

Solar-powered system offers a route to inexpensive desalination

Nurturing human communities and natural ecosystems