Michigan Engineering | University of Michigan Smarter use of processor speeds saves energy without compromising training speed and performance.
The post Up to 30% of the power used to train AI is wasted. Here’s how to fix it. appeared first on Michigan Engineering News.
Up to 30% of the power used to train AI is wasted. Here’s how to fix it.
-
Precision health and advanced communications: €9M ($10M) for bio-inspired nanoparticles on demand
Advanced microscopy techniques and AI models will help design complex nanoparticles for specific biological targets with less trial and error.
The post Precision health and advanced communications: €9M ($10M) for bio-inspired nanoparticles on demand appeared first on Michigan Engineering News.
-
Versatile knee exo for safer lifting
Helping out the quad muscles kept study participants lifting safely despite fatigue, with an algorithm that smoothly shifts between lifting and carrying tasks.
The post Versatile knee exo for safer lifting appeared first on Michigan Engineering News.
-
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.
-
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.
-
Outsmarting floodwaters
The world needs better tools to protect people during floods. A new machine learning method could help.
The post Outsmarting floodwaters appeared first on Michigan Engineering News.
-
Open source platform enables research on privacy-preserving machine learning
Virtual assortment of user devices provides a realistic training environment for distributed machine learning, protects privacy by learning where data lives.
The post Open source platform enables research on privacy-preserving machine learning appeared first on Michigan Engineering News.
-
Machine learning begins to understand the human gut
The new computer model accurately predicts the behavior of millions of microbial communities from hundreds of experiments, an advance toward precision medicine.
-
Update: Augmented reality engineering startup lands $1.1M SBIR grant
University of Michigan to explore inventive uses for the technology, including material science, biology and medicine
The post Update: Augmented reality engineering startup lands $1.1M SBIR grant appeared first on Michigan Engineering News.
-
Immune to hacks: Inoculating deep neural networks to thwart attacks
The adaptive immune system serves as a template for defending neural nets from confusion-sowing attacks