no code implementations • 29 Sep 2023 • Tuan Le, Julian Cremer, Frank Noé, Djork-Arné Clevert, Kristof Schütt
To further strengthen the applicability of diffusion models to limited training data, we investigate the transferability of EQGAT-diff trained on the large PubChem3D dataset with implicit hydrogen atoms to target different data distributions.
no code implementations • 3 Feb 2023 • Marco Bertolini, Van-Khoa Le, Jake Pencharz, Andreas Poehlmann, Djork-Arné Clevert, Santiago Villalba, Floriane Montanari
We validate quantitatively our methods by quantifying the agreements of our explanations' heatmaps with pathologists' annotations, as well as with predictions from a segmentation model trained on such annotations.
Explainable Artificial Intelligence (XAI) whole slide images
no code implementations • 20 Feb 2022 • Tuan Le, Frank Noé, Djork-Arné Clevert
Learning and reasoning about 3D molecular structures with varying size is an emerging and important challenge in machine learning and especially in drug discovery.
no code implementations • 18 Feb 2022 • Marco Bertolini, Djork-Arné Clevert, Floriane Montanari
Finally, we show that adopting our proposed scores as constraints during the training of a representation learning task improves the downstream performance of the model.
no code implementations • 15 Feb 2022 • Robin Winter, Marco Bertolini, Tuan Le, Frank Noé, Djork-Arné Clevert
In this work, we extend group invariant and equivariant representation learning to the field of unsupervised deep learning.
1 code implementation • 11 May 2021 • Ryan Henderson, Djork-Arné Clevert, Floriane Montanari
Rationalizing which parts of a molecule drive the predictions of a molecular graph convolutional neural network (GCNN) can be difficult.
1 code implementation • NeurIPS 2021 • Robin Winter, Frank Noé, Djork-Arné Clevert
In this work we address this issue by proposing a permutation-invariant variational autoencoder for graph structured data.
1 code implementation • 30 Mar 2021 • Tuan Le, Marco Bertolini, Frank Noé, Djork-Arné Clevert
Despite recent advances in representation learning in hypercomplex (HC) space, this subject is still vastly unexplored in the context of graphs.
Ranked #15 on Graph Property Prediction on ogbg-molpcba
no code implementations • 5 Jan 2021 • Robin Winter, Frank Noé, Djork-Arné Clevert
In this work we introduce an Autoencoder for molecular conformations.
no code implementations • 9 Oct 2020 • Ryan Henderson, Djork-Arné Clevert, Floriane Montanari
Due to the nature of deep learning approaches, it is inherently difficult to understand which aspects of a molecular graph drive the predictions of the network.
no code implementations • 25 Sep 2019 • Paul Kim, Robin Winter, Djork-Arné Clevert
We apply this reasoning to propose a novel proteochemometric modeling methodology which, for the first time, uses embeddings generated via unsupervised representation learning for both the protein and ligand descriptors.
no code implementations • ICLR 2018 • Robin Winter, Djork-Arné Clevert
Although GANs can learn a rich representation of the covered modes of the data in their latent space, the framework misses an inverse mapping from data to this latent space.
16 code implementations • 23 Nov 2015 • Djork-Arné Clevert, Thomas Unterthiner, Sepp Hochreiter
In contrast to ReLUs, ELUs have negative values which allows them to push mean unit activations closer to zero like batch normalization but with lower computational complexity.
Ranked #144 on Image Classification on CIFAR-100 (using extra training data)
no code implementations • NeurIPS 2015 • Djork-Arné Clevert, Andreas Mayr, Thomas Unterthiner, Sepp Hochreiter
We proof convergence and correctness of the RFN learning algorithm.