New technique overcomes spurious correlations problem in AI
Peer-Reviewed Publication
Updates every hour. Last Updated: 20-Aug-2025 17:09 ET (20-Aug-2025 21:09 GMT/UTC)
AI models often rely on “spurious correlations,” making decisions based on unimportant and potentially misleading information. Researchers have now discovered these learned spurious correlations can be traced to a very small subset of the training data and have demonstrated a technique that overcomes the problem.
In a new study published in Engineering, researchers from Huazhong University of Science and Technology and the Technical University of Munich have developed an improved proximal policy optimization (IPPO) method. This method is designed to solve the distributed heterogeneous hybrid blocking flow-shop scheduling problem (DHHBFSP), aiming to minimize total tardiness and total energy consumption. The research offers a practical approach for manufacturing scheduling, and its experimental results show better performance compared with other methods.
EPFL researchers have achieved a remarkable result: capturing and studying phase changes in quantum hardware, which hold hold promise for next-generation technologies like quantum computing and ultra-sensitive sensors.
In the field of wireless communication, security is of utmost importance. A new study published in Engineering explores intelligent covert communication. It reviews current techniques across different domains, from time and frequency to spatial and modulation. The research also looks at future trends like intelligent cooperative and parasitic covert communication, as well as challenges such as dealing with active detection and integrating sensing and communication. This offers valuable insights for the development of more secure wireless communication methods.
A recent review has shed light on the transformative potential of acoustic technologies in the fields of biofabrication and tissue engineering.
Quaise Energy reports new insights into designing superhot geothermal plants in two recent papers. The geothermal plants of today work at relatively low temperatures, most ranging from about 100 to 250oC. Less is known about designing plants for tapping into the heat in superhot rock over 375oC, which could offer significant benefits. Daniel Dichter of Quaise is working to understand more, and presented his most recent results at two recent conferences.