Related Experiment Video
Updated: Aug 13, 2025
05:30
Large Scale Energy Efficient Sensor Network Routing Using a Quantum Processor Unit
Published on: September 8, 2023
632
Bayesian chemical reaction neural network for autonomous kinetic uncertainty quantification
Qiaofeng Li1, Huaibo Chen1, Benjamin C Koenig1
1Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, MA 02139, USA. silideng@mit.edu.
Physical Chemistry Chemical Physics : PCCP
|January 20, 2023
Summary
Bayesian Chemical Reaction Neural Networks (CRNNs) now quantify uncertainty in chemical kinetic models. This data-driven approach integrates physical laws for robust and interpretable reaction discovery.
Area of Science:
- Chemical Engineering
- Computational Chemistry
- Biochemistry
Background:
- Chemical Reaction Neural Networks (CRNNs) are advanced tools for autonomous reaction model discovery.
- CRNNs combine deep neural networks (DNNs) with physical laws like mass action and Arrhenius law for interpretability and robustness.
Purpose of the Study:
- To develop Bayesian CRNN for quantifying uncertainty in chemical kinetic models derived from data.
- To explore methods for uncertainty quantification in data-driven chemical modeling.
Main Methods:
- Implemented Bayesian inference using Markov chain Monte Carlo (MCMC) and variational inference (VI).
- Utilized variational inference for its computational speed in Bayesian CRNN analysis.
- Applied Bayesian CRNN to various chemical systems for kinetic uncertainty quantification.
Main Results:
- Successfully demonstrated Bayesian CRNN's capability in quantifying kinetic uncertainty across different chemical systems.
- Highlighted the importance of integrating physical laws into data-driven models for enhanced reliability.
- Showcased adaptations for incomplete measurements and model mixing for comprehensive uncertainty quantification.
Conclusions:
- Bayesian CRNN provides a robust framework for uncertainty quantification in chemical kinetic modeling.
- Integrating physical laws is crucial for reliable data-driven chemical discovery.
- The method is adaptable for complex scenarios including incomplete data and model ensembles.
More Related Videos
10:44Inherent Dynamics Visualizer, an Interactive Application for Evaluating and Visualizing Outputs from a Gene Regulatory Network Inference Pipeline
Published on: December 7, 2021
2.2K
11:18Closed-loop Neuro-robotic Experiments to Test Computational Properties of Neuronal Networks
Published on: March 2, 2015
10.4K
Related Concept Videos
03:00Standard Entropy Change for a Reaction
20.7K
Entropy is a state function, so the standard entropy change for a chemical reaction (ΔS°rxn) can be calculated from the difference in standard entropy between the products and the reactants.
20.7K
01:24Chemical Reaction Rate
Chemical Reaction RateThe rate of a chemical reaction refers to how quickly or slowly a reaction occurs. It measures the change in the concentration of reactants or products over time. Some reactions, such as burning gasoline, happen very quickly, while others, like the rusting of iron, occur slowly over days or even weeks.Measuring Reaction Rate:The reaction rate can be measured usingThe disappearance of reactants over timeThe appearance of products over timeChanges in temperature or color of...
00:59Uncertainty: Overview
642
In analytical chemistry, we often perform repetitive measurements to detect and minimize inaccuracies caused by both determinate and indeterminate errors. Despite the cares we take, the presence of random errors means that repeated measurements almost never have exactly the same magnitude. The collective difference between these measurements - observed values - and the estimated or expected value is called uncertainty. Uncertainty is conventionally written after the estimated or expected value.
642
01:10Propagation of Uncertainty from Systematic Error
587
The atomic mass of an element varies due to the relative ratio of its isotopes. A sample's relative proportion of oxygen isotopes influences its average atomic mass. For instance, if we were to measure the atomic mass of oxygen from a sample, the mass would be a weighted average of the isotopic masses of oxygen in that sample. Since a single sample is not likely to perfectly reflect the true atomic mass of oxygen for all the molecules of oxygen on Earth, the mass we obtain from this...
587
02:24Predicting Reaction Outcomes
8.5K
Kinetics describes the rate and path by which a reaction occurs. In contrast, thermodynamics deals with state functions and describes the properties, behavior, and components of a system. It is not concerned with the path taken by the process and cannot address the rate at which a reaction occurs. Although it does provide information about what can happen during a reaction process, it does not describe the detailed steps of what appears on an atomic or a molecular level. On the other hand,...
8.5K
00:59Propagation of Uncertainty from Random Error
766
An experiment often consists of more than a single step. In this case, measurements at each step give rise to uncertainty. Because the measurements occur in successive steps, the uncertainty in one step necessarily contributes to that in the subsequent step. As we perform statistical analysis on these types of experiments, we must learn to account for the propagation of uncertainty from one step to the next. The propagation of uncertainty depends on the type of arithmetic operation performed on...
766