C
Chenru Duan
Researcher at Massachusetts Institute of Technology
Publications - 53
Citations - 1122
Chenru Duan is an academic researcher from Massachusetts Institute of Technology. The author has contributed to research in topics: Computer science & Medicine. The author has an hindex of 14, co-authored 33 publications receiving 627 citations. Previous affiliations of Chenru Duan include Singapore–MIT alliance & Zhejiang University.
Papers
More filters
Journal ArticleDOI
A quantitative uncertainty metric controls error in neural network-driven chemical discovery
TL;DR: In this paper, the authors introduce the distance to available data in the latent space of a neural network ML model as a low-cost, quantitative uncertainty metric that works for both inorganic and organic chemistry.
Journal ArticleDOI
Accurate Multiobjective Design in a Space of Millions of Transition Metal Complexes with Neural-Network-Driven Efficient Global Optimization.
TL;DR: The ANN-driven EI approach achieves at least 500-fold acceleration over random search, identifying a Pareto-optimal design in around 5 weeks instead of 50 years, and shows that a multitask ANN with latent-distance-based UQ surpasses the generalization performance of a GP in this space.
Journal ArticleDOI
Strategies and Software for Machine Learning Accelerated Discovery in Transition Metal Chemistry
TL;DR: In this paper, the authors compare the performance of LASSO, kernel ridge regression (KRR), and artificial neural network (ANN) models using heuristic, topological revised autocorrelation (RAC) descriptors.
Journal ArticleDOI
Zero-temperature localization in a sub-Ohmic spin-boson model investigated by an extended hierarchy equation of motion
TL;DR: In this article, the hierarchy equation of motion (HEOM) is extended to the zero-temperature sub-Ohmic spin-boson model, providing a numerically accurate prediction of quantum dynamics.
Journal ArticleDOI
Designing in the Face of Uncertainty: Exploiting Electronic Structure and Machine Learning Models for Discovery in Inorganic Chemistry.
TL;DR: Five key mandates for realizing computationally driven accelerated discovery in inorganic chemistry are outlined, including fully automated simulation of new compounds, knowledge of prediction sensitivity or accuracy, faster-than-fast property prediction methods, and maps for rapid chemical space traversal.