Physics-Embedded Bayesian Neural Network Developed for Predicting Energy-Dependent Fission Product Yields
Researchers have developed a physics-embedded Bayesian neural network (PE-BNN) framework that predicts fission product yields across different energy levels by integrating nuclear physics knowledge into the model. The approach incorporates an energy-independent phenomenological shell factor as an input feature and uses the Watanabe-Akaike Information Criterion for hyperparameter optimization. This method is significant for nuclear physics research as it demonstrates how physics-informed machine learning can improve predictions of complex nuclear phenomena while capturing both fine structures and global energy trends.
A new machine learning framework combining Bayesian neural networks with embedded nuclear physics knowledge has been developed to predict fission product yields (FPYs) as a function of energy. The PE-BNN approach integrates a phenomenological shell factor as a single input feature, allowing the model to capture both fine-scale structures and broader energy-dependent trends in fission data. The researchers optimized hyperparameters using the Watanabe-Akaike Information Criterion (WAIC) to enhance predictive performance. Results show close agreement with known shell effects and prompt neutron multiplicities, demonstrating the effectiveness of physics-informed inputs for modeling observables with systematic features. The framework appears particularly well-suited for nuclear physics applications where domain knowledge can be meaningfully embedded into machine learning architectures.
What different sources said
- arXiv physicsCenter
A physics-embedded Bayesian neural network for predicting the energy dependence of fission product yields with fine structures
Related
Topology-Aware Thermodynamics Improves DNA Probe Specificity Design
Researchers developed a new framework for designing DNA probes that accounts for the spatial organization of matched sequences, not just overall thermodynamic stability. Traditional methods rely on scalar measures like melting temperature and free energy, which miss how mismatches are distributed along the probe. The approach could improve diagnostic accuracy in applications like HPV detection and gene expression profiling.
Study Identifies Optimal Thermal Dose for Combining Focused Ultrasound with Immunotherapy in Tumors
Researchers used multimodal PET imaging to identify an optimal thermal dose range for focused ultrasound ablation that destroys tumor tissue while preserving conditions for immunotherapy delivery. The study found that excessive heating collapses blood vessels needed for antibody access, while insufficient heating fails to adequately reduce tumor burden. The findings could guide clinical design of combination treatments pairing thermal ablation with immunotherapies.
Plant MSH1 Protein Functions as Mismatch-Directed Nuclease for Organelle Genome Maintenance
Researchers have identified the precise mechanism by which the AtMSH1 protein in Arabidopsis plants recognizes and cleaves DNA mismatches and lesions, preventing mutations in organellar genomes. The protein combines a DNA mismatch recognition module with a nuclease domain that makes staggered cuts at specific positions relative to DNA damage. This discovery explains how plants maintain unusually low mutation rates in their mitochondrial and chloroplast DNA compared to other eukaryotes.