Search NASA⌕ Search

Engineering topics

Ruiz, Sergio

Publications and source records attributed to Ruiz, Sergio.

Automating neutron resonances classification with Machine Learning [Slides]

Team reported the following accomplishments: the development of a Machine-Learning method to properly assign spins to neutron resonances (automated, general, reproducible); full integration with evaluated resonances in the Atlas (automation of new editions); training and optimization in synthetic data; validation and deployment to real experimental resonances. Future perspectives include exploration of other classifiers and hyper-parameter combinations, further validation with well-known nucleus (e.g. 235 U), and publication pipeline.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

New Opportunities to Study Earthquake Precursors

The topic of earthquake prediction has a long history, littered with failed attempts. Part of the challenge is that possible precursory signals are usually reported after the event, and the systematic relationships between potential precursors and main events, should they exist, are unclear. Furthermore, several recent studies have shown the potential of new approaches to simultaneously detect earthquake foreshocks and slow-slip phenomena through ground deformation, seismic, and gravitational transients—weeks to months before large subduction zone earthquakes. The entire international community of earthquake researchers should be engaged in deploying instrumentation, sharing data in real time, and improving physical models to resolve the extent to which slow slip events and earthquake swarms enhance the likelihood (or not) for later, larger earthquakes.

58 GEOSCIENCES↗

A Hybrid Reinforcement Learning and Cellular Automata Model for Crowd Simulation on the GPU

We present a GPU-based hybrid model for crowd simulations. The model uses reinforcement learning to guide groups of pedestrians towards a goal while adapting to environmental dynamics, and a cellular automaton to describe individual pedestrians’ interactions. In contrast to traditional multi-agent reinforcement learning methods, our model encodes the learned navigation policy into a navigation map, which is used by the cellular automaton’s update rule to calculate the next simulation step. As a result, reinforcement learning is independent of the number of agents, allowing the simulation of large crowds. Implementation of this model on the GPU allows interactive simulations of several hundreds of pedestrians.

Ruiz, Sergio↗