qfx

A new deep learning approach, optimized by genetic algorithms, promises to sharpen our ability to detect and analyze neutrinos from nuclear reactors.

A new computer vision system automatically analyzes raw soccer footage to generate 2D field representations and unlock detailed insights into player movements and team strategies.

A new approach to reinforcement learning leverages physical principles to enable autonomous vehicles to master complex racing scenarios without relying on pre-built maps.

Researchers propose a novel Bayesian approach to valuing information and building scalable oversight mechanisms for training AI models with human feedback.

Researchers are leveraging self-generated, synthetic forgeries to train more robust deepfake detection models and overcome limitations in existing approaches.

A new study demonstrates that a traditional machine learning approach can effectively identify images created by diffusion models, rivaling the performance of deep neural networks.

New research details a method for generating large, realistic datasets by combining experimental data with process simulation, dramatically improving the performance of deep learning models for identifying unusual events in batch distillation.

Researchers are leveraging deep learning models to improve the identification and categorization of glitches in gravitational wave detector data, a critical step towards uncovering signals from the universe.

A new approach uses artificial intelligence to enable more efficient and sustainable energy trading between local microgrids.

A new system is automating research across multiple fields, from code optimization to machine learning, using the power of large language models.