Machine learning
4M: a next-generation framework for training multimodal foundation
An open-source training framework to advance multimodal AI.
Why text message reminders may fall short for medication refills?
The study is one of the largest and most diverse of its kind and reveals that digital strategies need to be tested.
Neutron star merger simulations contribute to train AI
Collaborative effort uses Laboratory simulation to train AI for scientific discovery.
Most recent Large Language Models remain vulnerable to simple manipulations
Can we convince AI to answer harmful requests?
Multimodal language and graph learning of adsorption configuration in catalysis
Multimodal machine learning model increases accuracy.
Researchers use AI to discover a record-breaking material for film capacitors
Researchers use an innovative set of machine-learning models to discover a record-breaking material for film capacitors.
A new photonic chip for ultrafast AI computations
This new device uses light to perform the key operations of a deep neural network on a chip.
Machine learning helps find promising compositions for sodium-ion batteries
Researchers optimize the composition of a multi-element transition metal oxide.
AI algorithm accurately detects heart murmurs in dogs
In tests, the algorithm detected heart murmurs with a sensitivity of 90%.
New machine learning model predicts dielectric function of materials
The model speeds up dielectric materials development as a step toward 6G networks.
Computer simulations help pave the way for more efficient solar cells
Researchers have gained new insights into how perovskite materials function.
Deep-learning model to’ reverse-engineer’ new composite materials
Deep-learning model that can customize microarchitecture based on specific needs.
Machine learning uncovers rheumatoid arthritis subtypes
The study shows that machine learning can effectively identify subtypes of rheumatoid arthritis, helping doctors.
The rising extinction risk of marine Teleost fish species
The "silent extinction" of 1,337 threatened species includes fish families crucial to reef ecosystems.
AI predicts cycle lives of high-energy-density lithium-metal batteries
Major advances in putting the batteries into practical use.
Machine learning and multiple sclerosis progression
Can a computer predict disability progression in multiple sclerosis?
Machine learning could help identify plasmoids in outer space
The technology could aid efforts to answer long-standing astrophysical questions.
AI technology to predict yield strength of various metals
The Future of Metals Research with Artificial Intelligence.
Researchers develop new efficient, environmentally friendly deicer
New deicing mixture offers higher performance than deicers on the market.
An objective, noninvasive, and zero-effort sleep monitoring system using smart thermostats
A new study found a novel use for a smart household device.
Climate change and its impact on North American animal species
North American cities may see a significant species turnover by the end of the century.
EPFL robot notably learned to transition from trotting to pronking
Trotting robots reveal emergence of animal gait transitions.
Machine learning enhances vertical-axis wind turbine performance
They used a genetic learning algorithm to determine the best pitch for a VAWT blade.
A new method calculates the mechanical properties of solids using machine learning
The new method also benefits from active learning on local atomic environments.
Enabling AI models to see the world more like humans do
Researchers enhance peripheral vision in AI models.
Several air contaminants are linked to children’s asthma symptoms
It’s not just one pollutant that can be linked to asthma outcomes.