PRX Intelligence

PRX Intelligence Scope

PRX Intelligence welcomes manuscripts on AI/ML methods as well as their use across the physical sciences and related fields. Coverage spans theory, computation, and experiment in conventional physics disciplines and related domains—including computer science and mathematics, materials science and engineering, chemistry and biology, earth and environmental sciences—whenever AI/ML advances a physics based scientific understanding or capability.

Representative topics include, but are not limited to:

• AI/ML methods for science: physics-informed learning, surrogate and hybrid models, generative models for scientific data, probabilistic modeling, uncertainty quantification, interpretability and reliability, causal and mechanistic machine learning, reinforcement learning for design and control, scientific agents and automation, AI based optimization, statistical learning applied to experimental planning and system engineering.

• Data, simulation, and modeling: multimodal experimental, simulated, or synthetic datasets, meta data assimilation and curation from literature, scientific workload, cost and timing data.

• Domain applications: condensed-matter and materials, discovery and synthesis, soft matter and fluids, plasma and fusion, atomic/molecular/optical physics, high-energy and nuclear physics, astrophysics and cosmology, geophysics and climate, energy systems and devices, imaging and sensing (including microscopy and tomography).

• Systems, tools, and instrumentation: ML-enabled experimental design and control, robotics for laboratories, adaptive/streaming analyses at facilities, edge/embedded ML for instruments, ML pipelines integrated with experimental or observational platforms, high-performance computing.

• Emerging areas: quantum-aware/quantum-enhanced ML and ML for quantum systems, ML-guided theorem proving and scientific reasoning, trustworthy and robust and safety-aware ML for scientific deployment, agentic and robotic AI.