2022 Accepted Works

Behind the Scenes

• Hyperparameter Optimization of Graph Neural Networks for the OpenCatalyst Dataset: A Case Study

• Multivariate Prediction Intervals for Random Forests

• Human-in-the-Loop Approaches For Task Guidance In Manufacturing Settings

• Experimental platform and digital twin for AI-driven materials optimization and discovery for microelectronics using atomic layer deposition

• On Multi-information source Constraint Active Search

• Geometric Considerations for Normalization Layers in Equivariant Neural Networks

Papers

• A deep learning and data archaeology approach for mosquito repellent discovery

• Conformer Search Using SE3-Transformers and Imitation Learning

• AI-assisted chemical reaction impurity prediction and propagation

• Neural Structure Fields with Application to Crystal Structure Autoencoders

• Extracting Structural Motifs from Pair Distribution Function Data of Nanostructures using Explainable Machine Learning

• Group SELFIES: A Robust Fragment-Based Molecular String Representation

• Graph Contrastive Learning for Materials

• A Generalized Framework for Microstructural Optimization using Neural Networks

• DeepStruc: Towards structure solution from pair distribution function data using deep generative models

• A Data-efficient Multiobjective Machine Learning Method For 3D-printed Architected Materials Design

• Transfer Learning Lithium and Electrolyte Potential Energy Surfaces from Pure and Hybrid DFT

• Element-Wise Formulation of Inorganic Retrosynthesis

• MEGAN: Multi Explanation Graph Attention Network

• Accelerating the Discovery of Rare Materials with Bounded Optimization Techniques

• Deep Reinforcement Learning for Inverse Inorganic Materials Design

• A High-Throughput Platform for Efficient Exploration of Polypeptides Chemical Space via Automation and Machine Learning

• Robust design of semi-automated clustering models for 4D-STEM datasets

• Actively Learning Costly Reward Functions for Reinforcement Learning

• More trustworthy Bayesian optimization of materials properties by adding human into the loop

• Is GPT-3 all you need for machine learning for chemistry?

• The Largest Knowledge Graph in Materials Science - Entities, Relations, and Link Prediction through Graph Representation Learning

• Differential top-k learning for template-based single-step retrosynthesis

• PhAST: Physics-Aware, Scalable, and Task-specific GNNs for Accelerated Catalyst Design

• Self-driving Multimodal Studies at User Facilities

• Information Recovery via Matrix Completion for Piezoresponse Force Microscopy Data

• A Self-driving Laboratory Optimizes A Scalable Process For Making Functional Coatings

• Assessing multi-objective optimization of molecules with genetic algorithms against relevant baselines

• Generative Design of Material Microstructures for Organic Solar Cells using Diffusion Models

• Multi-Objective GFlowNets

• Integrating AI, automation and multiscale simulations for end-to-end design of phase-separating proteins

Proposals

• A Survey on Evaluation Metrics for Synthetic Material Micro-Structure Images from Generative Models

• Autonomous Materials Discovery for Organic Photovoltaics

Software & Tutorials

• Open MatSci ML Toolkit: A Flexible Framework for Machine Learning in Materials Science

• MolPAL: Software for Sample Efficient High-Throughput Virtual Screening