In the rapidly evolving field of machine learning and artificial intelligence, it’s challenging to stay up-to-date with the latest discoveries and innovations. This column, Perceptron (formerly Deep Science), aims to collect and explain the most relevant recent research papers and breakthroughs in AI and related fields.
Artificial Skin that Can Learn and React
A team of engineers at the University of Glasgow has developed "artificial skin" that can learn to experience and react to simulated pain. This breakthrough leverages a new type of processing system based on "synaptic transistors" designed to mimic the brain’s neural pathways.
The transistors, made from zinc-oxide nanowires printed onto the surface of a flexible plastic, connected to a skin sensor that registered changes in electrical resistance. Unlike previous attempts at artificial skin, this design used a circuit built into the system to act as an "artificial synapse," reducing input to a spike in voltage.
This innovation allows the team to "teach" the skin how to respond to simulated pain by setting a threshold of input voltage whose frequency varied according to the level of pressure applied to the skin. The potential applications for this technology are vast, including robotics, where it could prevent robotic arms from coming into contact with dangerously high temperatures.
Predicting Soccer Player Movement
DeepMind has developed an AI model called Graph Imputer that can anticipate where soccer players will move on the field. This breakthrough leverages graph neural networks to analyze player movement patterns and predict future locations.
The model is trained on a large dataset of soccer matches, allowing it to learn complex relationships between players’ movements and the game’s dynamics. The potential applications for this technology are numerous, including improving player tracking, tactical analysis, and fan engagement.
AI-Powered Clinical Diagnosis
Researchers at the Technical University of Munich have developed a clinical meta-algorithm that integrates multiple data sources (including other algorithms) to differentiate between certain liver diseases with similar presentations.
This breakthrough leverages machine learning to analyze complex patterns in patient data, allowing clinicians to identify subtle differences between conditions. While this technology won’t replace human doctors, it will continue to help wrangle the growing volumes of data that even specialists may not have the time or expertise to interpret.
AI-Powered Molecular Signature Analysis
A team of researchers has developed an AI model that can analyze molecular signatures to identify toxic chemicals and diagnose disease. This breakthrough leverages machine learning to learn complex patterns in molecular data, allowing it to outperform other AI models in identifying toxic chemicals in a test database.
The potential applications for this technology are vast, including environmental monitoring, clinical diagnosis, and personalized medicine.
Other Notable Breakthroughs
- Researchers have developed an AI model that can analyze brain activity to predict language development in children.
- A team of engineers has created a robotic arm that can learn to manipulate objects through machine learning.
- Scientists have discovered a new class of materials with potential applications in energy storage and electronics.
Conclusion
The field of AI is rapidly evolving, with breakthroughs emerging almost daily. This column aims to provide a snapshot of the most significant recent innovations and discoveries. Whether it’s artificial skin that can learn and react or AI-powered clinical diagnosis, these advancements have the potential to transform industries and improve lives.
Subscribe to TechCrunch AITechCrunch Daily News
Stay up-to-date with the latest AI news and breakthroughs by subscribing to our newsletters. Get the best of TechCrunch’s coverage delivered right to your inbox.
Related Articles
- Apple begins requiring JavaScript for Google Search
- Malware stole internal PowerSchool passwords from engineer’s hacked computer
- Trump comments on TikTok ban: ‘stay tuned’