Skip to Main content Open mobile menu Close mobile menu
Widely used AI tool for early sepsis detection may be cribbing doctors’ suspicions
  1. Widely used AI tool for early sepsis detection may be cribbing doctors’ suspicions

    When using only data collected before patients with sepsis received treatments or medical tests, the model’s accuracy was no better than a coin toss.

    The post Widely used AI tool for early sepsis detection may be cribbing doctors’ suspicions appeared first on Michigan Engineering News.

  2. Clinicians could be fooled by biased AI, despite explanations

    Regulators pinned their hopes on clinicians being able to spot flaws in explanations of an AI model’s logic, but a study suggests this isn’t a safe approach.

    The post Clinicians could be fooled by biased AI, despite explanations appeared first on Michigan Engineering News.

  3. Understanding attention in large language models

    How do chatbots based on the transformer architecture decide what to pay attention to in a conversation? They’ve made their own machine learning algorithms to tell them.

    The post Understanding attention in large language models appeared first on Michigan Engineering News.

  4. Biases in large image-text AI model favor wealthier, Western perspectives

    AI model that pairs text, images performs poorly on lower-income or non-Western images, potentially increasing inequality in digital technology representation.

    The post Biases in large image-text AI model favor wealthier, Western perspectives appeared first on Michigan Engineering News.

  5. Building a chemical ‘GPT’ to help design a key battery component

    Taking inspiration from the word-predicting large language models, a U-M team is kickstarting an atom-predicting model with 200,000 node hours on Argonne’s Polaris.

    The post Building a chemical ‘GPT’ to help design a key battery component appeared first on Michigan Engineering News.

  6. Choosing exoskeleton settings like a Pandora radio station

    Using a simple and convenient touchscreen interface, the algorithm learns the assistance preferences of the wearer.

    The post Choosing exoskeleton settings like a Pandora radio station appeared first on Michigan Engineering News.

  7. Toyota Research Institute and Michigan Engineering discuss importance of long-standing partnership

    As the auto industry expands beyond transportation, university collaborations are more important than ever.

    The post Toyota Research Institute and Michigan Engineering discuss importance of long-standing partnership appeared first on Michigan Engineering News.

  8. Senate testimony: AI risks to the financial sector

    U-M professor of computer science and engineering identifies new concerns that recent AI advances pose for financial markets.

    The post Senate testimony: AI risks to the financial sector appeared first on Michigan Engineering News.

  9. AI tool helps optimize antibody medicines

    Machine learning points out why antibodies fail to stay on target, binding to molecules that aren’t markers of disease—and suggests better designs.

    The post AI tool helps optimize antibody medicines appeared first on Michigan Engineering News.

  10. Nanobiotics: AI for discovering where and how nanoparticles bind with proteins

    A new tool in the fight against superbugs goes beyond protein folding simulations like AlphaFold, potentially revealing antibiotic candidates.

    The post Nanobiotics: AI for discovering where and how nanoparticles bind with proteins appeared first on Michigan Engineering News.