Search Results for author: Pascal Friederich

Found 14 papers, 9 papers with code

Neural networks trained on synthetically generated crystals can extract structural information from ICSD powder X-ray diffractograms

1 code implementation21 Mar 2023 Henrik Schopmans, Patrick Reiser, Pascal Friederich

However, training directly on simulated diffractograms from databases such as the ICSD is challenging due to its limited size, class-inhomogeneity, and bias toward certain structure types.

Connectivity Optimized Nested Graph Networks for Crystal Structures

1 code implementation27 Feb 2023 Robin Ruff, Patrick Reiser, Jan Stühmer, Pascal Friederich

Graph neural networks (GNNs) have been applied to a large variety of applications in materials science and chemistry.

graph construction

Actively Learning Costly Reward Functions for Reinforcement Learning

1 code implementation23 Nov 2022 André Eberhard, Houssam Metni, Georg Fahland, Alexander Stroh, Pascal Friederich

Transfer of recent advances in deep reinforcement learning to real-world applications is hindered by high data demands and thus low efficiency and scalability.

Active Learning reinforcement-learning +1

MEGAN: Multi-Explanation Graph Attention Network

1 code implementation23 Nov 2022 Jonas Teufel, Luca Torresi, Patrick Reiser, Pascal Friederich

However, existing explanation methods have limited expressiveness and interoperability due to the fact that only single explanations in form of node and edge importance are generated.

Explainable artificial intelligence Graph Attention +1

Graph neural networks for materials science and chemistry

no code implementations5 Aug 2022 Patrick Reiser, Marlen Neubert, André Eberhard, Luca Torresi, Chen Zhou, Chen Shao, Houssam Metni, Clint van Hoesel, Henrik Schopmans, Timo Sommer, Pascal Friederich

Machine learning plays an increasingly important role in many areas of chemistry and materials science, e. g. to predict materials properties, to accelerate simulations, to design new materials, and to predict synthesis routes of new materials.

On scientific understanding with artificial intelligence

no code implementations4 Apr 2022 Mario Krenn, Robert Pollice, Si Yue Guo, Matteo Aldeghi, Alba Cervera-Lierta, Pascal Friederich, Gabriel dos Passos Gomes, Florian Häse, Adrian Jinich, AkshatKumar Nigam, Zhenpeng Yao, Alán Aspuru-Guzik

Imagine an oracle that correctly predicts the outcome of every particle physics experiment, the products of every chemical reaction, or the function of every protein.


Implementing graph neural networks with TensorFlow-Keras

1 code implementation7 Mar 2021 Patrick Reiser, Andre Eberhard, Pascal Friederich

Graph neural networks are a versatile machine learning architecture that received a lot of attention recently.

Analyzing dynamical disorder for charge transport in organic semiconductors via machine learning

no code implementations2 Feb 2021 Patrick Reiser, Manuel Konrad, Artem Fediai, Salvador Léon, Wolfgang Wenzel, Pascal Friederich

Organic semiconductors are indispensable for today's display technologies in form of organic light emitting diodes (OLEDs) and further optoelectronic applications.

BIG-bench Machine Learning

Scientific intuition inspired by machine learning generated hypotheses

no code implementations27 Oct 2020 Pascal Friederich, Mario Krenn, Isaac Tamblyn, Alan Aspuru-Guzik

Machine learning with application to questions in the physical sciences has become a widely used tool, successfully applied to classification, regression and optimization tasks in many areas.

BIG-bench Machine Learning

Neural Message Passing on High Order Paths

no code implementations24 Feb 2020 Daniel Flam-Shepherd, Tony Wu, Pascal Friederich, Alan Aspuru-Guzik

Graph neural network have achieved impressive results in predicting molecular properties, but they do not directly account for local and hidden structures in the graph such as functional groups and molecular geometry.

Molecular Property Prediction

Self-Referencing Embedded Strings (SELFIES): A 100% robust molecular string representation

2 code implementations31 May 2019 Mario Krenn, Florian Häse, AkshatKumar Nigam, Pascal Friederich, Alán Aspuru-Guzik

SELFIES can be directly applied in arbitrary machine learning models without the adaptation of the models; each of the generated molecule candidates is valid.

Cannot find the paper you are looking for? You can Submit a new open access paper.