Humans are usually pretty good at recognising when they get things wrong, but artificial intelligence systems are not. According to a new study, AI generally suffers from inherent limitations due to a century-old mathematical paradox.
Humans are usually pretty good at recognising when they get things wrong, but artificial intelligence systems are not. According to a new study, AI generally suffers from inherent limitations due to a century-old mathematical paradox.
Designed to bring machine learning to the Earth's last great frontier, a prototype smart sensor platform users on-device inference to classify marine mammal calls — powering itself exclusively from the sounds it captures.
Artificial intelligence (AI) is a wide-ranging tool that enables people to rethink how we integrate information, analyze data, and use the resulting insights to improve decision making
"MechStyle" allows users to personalize 3D models, while ensuring they're physically viable after fabrication, producing unique personal items and assistive technology.
A simulation of an individual patient's brain tumors, kept up to date with readily available data, can identify whether dietary treatments and drugs are likely to work.
Hey Tuya showcases how agentic Physical AI evolves assistants from isolated commands into systems that learn context, coordinate devices, and act reliably across real-world environments.
This episode celebrates five years of The Next Byte with a recap of 2025, highlighting standout episodes through The "Saucies" Awards, reflecting on major trends, sharing predictions for 2026, and thanking the global listener community.
Matroid builds no-code computer-vision detectors that can spot everything from microscopic material defects to real-time safety hazards on a factory floor.
In large-scale warehousing and distribution operations, conveyor belts are an essential infrastructure that must operate with near-zero downtime to ensure the timely delivery of products. The presence of loose or foreign items on a conveyor belt can pose a serious risk to these operations.
In this post, we'll walk through how to evaluate that progress using the same metrics our platform provides automatically, so you can build detectors that get smarter, sharper, and more reliable over time.
Humans are usually pretty good at recognising when they get things wrong, but artificial intelligence systems are not. According to a new study, AI generally suffers from inherent limitations due to a century-old mathematical paradox.
Designed to bring machine learning to the Earth's last great frontier, a prototype smart sensor platform users on-device inference to classify marine mammal calls — powering itself exclusively from the sounds it captures.
For the second consecutive year, we have compiled a list of Campus companies that Eindhoven, the Netherlands and the world will hear more about in 2022. This year, we’re focusing on companies created in High Tech Campus’s startup ecosystem, or that landed on the Campus during 2021.
Research from MIT suggests that a certain type of robust computer-vision model perceives visual representations similarly to the way humans do using peripheral vision.
This 3D vision system for a mobile robot picker consists of an AMR with a robotic arm equipped with the Photoneo 3D camera MotionCam-3D. The hand-eye setup allows you to scan and approach each object from an appropriate angle, distance, and position.
Designed as a necklace-style wearable in a 3D-printed case, SpeeChin is capable of recognizing 54 English and 44 Chinese voice commands — even if the wearer never says anything out loud.
The in-sensor adaptation strategy widens the range for image perception under different illumination conditions to simplify the complexity of hardware and algorithms.
Sounds provide important information about how well a machine is running. ETH researchers have now developed a new machine learning method that automatically detects whether a machine is "healthy" or requires maintenance.
Designed to address the OOD data false prediction by adaptively synthesizing virtual outliers that can maintain the model’s decision boundary during training.
Designed to address the exploding computational demands of deep neural networks, physical neural networks (PNNs) branch out from electronics into optics and even mechanics to boost performance and efficiency.
Designed for use in the human body for everything from drug delivery to less-invasive biopsies, these tiny microrobots operate under the control of a deep-learning system trained with no modelling or prior environmental knowledge.