Machine learning

8 items, newest first.

Mar 2026
Talk
ML-Guided Ab initio Modeling
Seminar, LPENS, École Normale Supérieure, Paris
AI/ML Physics Machine learning DFT
2025
Publication
Machine Learning-Driven First-Principles Modeling of Far-Infrared Materials
Ritwik Das. PhD thesis, Université Paris-Saclay, C2N (CNRS). NNT 2025UPAST100
Physics AI/ML Bayesian optimization Gaussian processes Machine learning DFT+U Hybrid functionals GW InSb InAsSb alloys Heterointerfaces Band alignment Virtual crystal approximation Alloys Infrared materials Semiconductors
PhD, completed 2025
Project
Machine-learning-driven modeling of far-infrared semiconductors
InSb, InAs, InAsSb alloys and CdTe/InSb heterointerfaces from first principles.
Physics Electronics Semiconductors Infrared materials InSb InAsSb alloys Heterointerfaces Alloys DFT DFT+U Hybrid functionals GW DFPT Band alignment Bayesian optimization Machine learning
2025
Project
Deep-learning prediction of ab initio parameters
A pipeline, from an MLP to a graph neural network, that infers Hubbard U and related parameters from atomic and electronic descriptors.
AI/ML Physics Graph neural networks Machine learning DFT+U Hybrid functionals
2025
Publication
High-fidelity electronic structure and properties of InSb: G0W0 and Bayesian-optimized hybrid functionals and DFT + U approaches
Ritwik Das, Anne-Sophie Grimault-Jacquin, Frédéric Aniel. Physical Review B 112, 075136
Physics AI/ML InSb Semiconductors Infrared materials GW Hybrid functionals DFT+U Bayesian optimization Machine learning
2024
Publication
BMach: a Bayesian machine for optimizing Hubbard U parameters in DFT+U with machine learning
R. Das. arXiv preprint
AI/ML Physics Bayesian optimization Gaussian processes Machine learning DFT+U
2023
Talk
Bayesian Machine: Optimizing the Hubbard U Parameter in DFT+U With Machine Learning
Conference contribution, European Materials Research Society (E-MRS) Conference
AI/ML Physics Bayesian optimization Machine learning DFT+U
2023–2024
Project
BMach, Bayesian optimization of Hubbard U
A Bayesian framework that calibrates Hubbard U and hybrid-functional parameters at a small fraction of the cost of direct high-level methods.
AI/ML Physics Bayesian optimization Gaussian processes Machine learning DFT+U Hybrid functionals