Towards Self-Consistent Graph Neural Networks for Predicting the Ideal Gas Heat Capacity, Enthalpy, and Entropy
Talk, ESCAPE-33, Florence, Italy
I gave a talk about the work we produced here
Talks and presentations
State-aware graph neural networks for temperature-dependent propertie
Talk, ESCAPE-33, Athens, Greece
I gave a talk about the work we produced here
Combining Functional Groups and Graph Neural Networks Towards Interpretable Molecular Property Models
Talk, AIChE annual meeting 2022, Pheonix, USA
The abstract of the talk can be found here
Integrating Chemistry Knowledge and Graph Neural Networks Towards Interpretable Molecular Property Models
Poster, DDSA-2022, Billund, Denmark
I gave a talk about our working in combining chemistry knwoledge into graph neural networks.
DeepGSA: Plant Data-Driven Global Sensitivity Analysis using Deep Learning
Poster, ESCAPE-32, Kyoto, Japan
I gave a talk about the work we produced here
Molecular Representations in Deep-Learning Models for Chemical Property Prediction
Poster, ESCAPE-32, Kyoto, Japan
I gave a talk about the work we produced here
Application of Outlier Treatment Towards Improved Property Prediction Models
Poster, ESCAPE-32, Toulouse, France
I gave a talk about the work we produced here
An Interpretable Graph Neural Network-based Property Prediction
Talk, ESCAPE-32, Toulouse, France
I gave a talk about the work we produced here
Prediction of Nitrous Oxide Concentration in Wastewater Treatment Plants: Application and Benchmarking of Deep-Learning Models
Poster, AIChE Annual meeting 2021, Boston, USA
Abstract can be found here
Multi-Task Property Prediction: Importance of the “Chemist-in-the-Loop” in Model
Poster, AIChE Annual meeting 2021, Boston, USA
Abstract can be found here
Benchmarking Uncertainty Quantification Methods for Property Prediction Models: Application to Group-Contribution Models
Keynote lecture, the 13th European Congress of Chemical Engineering (ECCE), Berlin, Germany
I gave a keynote talk about benchmarking various uncertainty estimation techniques applied to group-contribution models.
Data Scarcity in Developing Property Prediction Models: Application of Multi-Task Transfer
Poster, 21st Symposium on thermophysical properties, Colorado, USA
I gave a talk on how to leverage multi-task and transfer learning in cases where the data availability is limited.
Comparison of Group-Contribution and Machine Learning-based Property Prediction Models with Uncertainty Quantification
Poster, ESCAPE-31, Istanbul, Turkey
I gave a talk concerning the work we produced here.
Importance of Outlier Treatment for Property Prediction Models
Talk, CAPE Forum 2020, Lyngby, Denmark
I gave a talk about the work we produced here