Search Results for author:
Found 82 papers, 43 papers with code
Winning the Lottery Ahead of Time: Efficient Early Network Pruning
Pruning, the task of sparsifying deep neural networks, received increasing attention recently.
Disentangling Epistemic and Aleatoric Uncertainty in Reinforcement Learning
Characterizing aleatoric and epistemic uncertainty can be used to speed up learning in a training environment, improve generalization to similar testing environments, and flag unfamiliar behavior in anomalous testing environments.
Sampling-free Inference for Ab-Initio Potential Energy Surface Networks
In recent work, the potential energy surface network (PESNet) has been proposed to reduce training time by solving the Schr\"odinger equation for many geometries simultaneously.
Predicting single-cell perturbation responses for unseen drugs
Single-cell transcriptomics enabled the study of cellular heterogeneity in response to perturbations at the resolution of individual cells.
How Do Graph Networks Generalize to Large and Diverse Molecular Systems?
Based on this analysis, we identify a smaller dataset that correlates well with the full OC20 dataset, and propose the GemNet-OC model, which outperforms the previous state-of-the-art on OC20 by 16%, while reducing training time by a factor of 10.
Ranked #1 on
Initial Structure to Relaxed Energy (IS2RE)
on OC20
Learning the Dynamics of Physical Systems from Sparse Observations with Finite Element Networks
We propose a new method for spatio-temporal forecasting on arbitrarily distributed points.
Interpretability Techniques for Deep Learning
Interpretable Machine Learning
+2
Differentiable DAG Sampling
To this end, DP-DAG samples a DAG by successively (1) sampling a linear ordering of the node and (2) sampling edges consistent with the sampled linear ordering.
Is it all a cluster game? -- Exploring Out-of-Distribution Detection based on Clustering in the Embedding Space
It is essential for safety-critical applications of deep neural networks to determine when new inputs are significantly different from the training distribution.
Enabling Automated Machine Learning for Model-Driven AI Engineering
Developing smart software services requires both Software Engineering and Artificial Intelligence (AI) skills.
Graph Data Augmentation for Graph Machine Learning: A Survey
In this paper, we present a comprehensive and systematic survey of graph data augmentation that summarizes the literature in a structured manner.
Multi-Objective Model Selection for Time Series Forecasting
By learning a mapping from forecasting models to performance metrics, we show that our method PARETOSELECT is able to accurately select models from the Pareto front -- alleviating the need to train or evaluate many forecasting models for model selection.
Directional Message Passing on Molecular Graphs via Synthetic Coordinates
We furthermore set the state of the art on ZINC and coordinate-free QM9 by incorporating synthetic coordinates in the SMP and DimeNet++ models.
Robustness of Graph Neural Networks at Scale
Graph Neural Networks (GNNs) are increasingly important given their popularity and the diversity of applications.
Graph Posterior Network: Bayesian Predictive Uncertainty for Node Classification
GPN outperforms existing approaches for uncertainty estimation in the experiments.
Neural Flows: Efficient Alternative to Neural ODEs
Neural ordinary differential equations describe how values change in time.
Generalization of Neural Combinatorial Solvers Through the Lens of Adversarial Robustness
Specifically, most datasets only capture a simpler subproblem and likely suffer from spurious features.
Intriguing Properties of Input-dependent Randomized Smoothing
We show that in general, the input-dependent smoothing suffers from the curse of dimensionality, forcing the variance function to have low semi-elasticity.
Ab-Initio Potential Energy Surfaces by Pairing GNNs with Neural Wave Functions
Solving the Schr\"odinger equation is key to many quantum mechanical properties.
3D Infomax improves GNNs for Molecular Property Prediction
Molecular property prediction is one of the fastest-growing applications of deep learning with critical real-world impacts.
Localized Randomized Smoothing for Collective Robustness Certification
Models for image segmentation, node classification and many other tasks map a single input to multiple labels.
3D Pre-training improves GNNs for Molecular Property Prediction
Molecular property prediction is one of the fastest-growing applications of deep learning with critical real-world impacts.
End-to-End Learning of Probabilistic Hierarchies on Graphs
We propose a novel probabilistic model over hierarchies on graphs obtained by continuous relaxation of tree-based hierarchies.
Locality-Based Mini Batching for Graph Neural Networks
Training graph neural networks on large graphs is challenging since there is no clear way of how to extract mini batches from connected data.
Provably Robust Transfer
Our experiments show how adversarial training on the source domain affects robustness on source and target domain, and we propose the first provably robust transfer learning models.
A Study of Joint Graph Inference and Forecasting
We study a recent class of models which uses graph neural networks (GNNs) to improve forecasting in multivariate time series.
On Second-order Optimization Methods for Federated Learning
We consider federated learning (FL), where the training data is distributed across a large number of clients.
Whole Brain Vessel Graphs: A Dataset and Benchmark for Graph Learning and Neuroscience (VesselGraph)
Moreover, we benchmark numerous state-of-the-art graph learning algorithms on the biologically relevant tasks of vessel prediction and vessel classification using the introduced vessel graph dataset.
OODformer: Out-Of-Distribution Detection Transformer
A serious problem in image classification is that a trained model might perform well for input data that originates from the same distribution as the data available for model training, but performs much worse for out-of-distribution (OOD) samples.
Scalable Optimal Transport in High Dimensions for Graph Distances, Embedding Alignment, and More
The current best practice for computing optimal transport (OT) is via entropy regularization and Sinkhorn iterations.
MDE4QAI: Towards Model-Driven Engineering for Quantum Artificial Intelligence
Over the past decade, Artificial Intelligence (AI) has provided enormous new possibilities and opportunities, but also new demands and requirements for software systems.
Graphhopper: Multi-Hop Scene Graph Reasoning for Visual Question Answering
We conduct an experimental study on the challenging dataset GQA, based on both manually curated and automatically generated scene graphs.
Supporting AI Engineering on the IoT Edge through Model-Driven TinyML
We focus on a sub-discipline of AI, namely Machine Learning (ML) and propose the delegation of data analytics and ML to the IoT edge.
ML-Quadrat & DriotData: A Model-Driven Engineering Tool and a Low-Code Platform for Smart IoT Services
In this paper, we present ML-Quadrat, an open-source research prototype that is based on the Eclipse Modeling Framework (EMF) and the state of the art in the literature of Model-Driven Software Engineering (MDSE) for smart Cyber-Physical Systems (CPS) and the Internet of Things (IoT).
A Model-Driven Approach to Machine Learning and Software Modeling for the IoT
In particular, we implement the proposed approach, called ML-Quadrat, based on ThingML, and validate it using a case study from the IoT domain, as well as through an empirical user evaluation.
On Out-of-distribution Detection with Energy-based Models
Several density estimation methods have shown to fail to detect out-of-distribution (OOD) samples by assigning higher likelihoods to anomalous data.
Detecting Anomalous Event Sequences with Temporal Point Processes
Automatically detecting anomalies in event data can provide substantial value in domains such as healthcare, DevOps, and information security.
GemNet: Universal Directional Graph Neural Networks for Molecules
Effectively predicting molecular interactions has the potential to accelerate molecular dynamics by multiple orders of magnitude and thus revolutionize chemical simulations.
Natural Posterior Network: Deep Bayesian Uncertainty for Exponential Family Distributions
Uncertainty awareness is crucial to develop reliable machine learning models.
Neural Temporal Point Processes: A Review
Temporal point processes (TPP) are probabilistic generative models for continuous-time event sequences.
Language-Agnostic Representation Learning of Source Code from Structure and Context
Source code (Context) and its parsed abstract syntax tree (AST; Structure) are two complementary representations of the same computer program.
Collective Robustness Certificates
In tasks like node classification, image segmentation, and named-entity recognition we have a classifier that simultaneously outputs multiple predictions (a vector of labels) based on a single input, i. e. a single graph, image, or document respectively.
Warpspeed Computation of Optimal Transport, Graph Distances, and Embedding Alignment
Optimal transport (OT) is a cornerstone of many machine learning tasks.
Deep Rao-Blackwellised Particle Filters for Time Series Forecasting
We propose a Monte Carlo objective that leverages the conditional linearity by computing the corresponding conditional expectations in closed-form and a suitable proposal distribution that is factorised similarly to the optimal proposal distribution.
Fast and Uncertainty-Aware Directional Message Passing for Non-Equilibrium Molecules
Many important tasks in chemistry revolve around molecules during reactions.
Reliable Graph Neural Networks via Robust Aggregation
Perturbations targeting the graph structure have proven to be extremely effective in reducing the performance of Graph Neural Networks (GNNs), and traditional defenses such as adversarial training do not seem to be able to improve robustness.
Evaluating Robustness of Predictive Uncertainty Estimation: Are Dirichlet-based Models Reliable?
Dirichlet-based uncertainty (DBU) models are a recent and promising class of uncertainty-aware models.
Scalable Normalizing Flows for Permutation Invariant Densities
Modeling sets is an important problem in machine learning since this type of data can be found in many domains.
Equivariant Normalizing Flows for Point Processes and Sets
To model this behavior, it is enough to transform the samples from the uniform process with a sufficiently complex equivariant function.
From Things' Modeling Language (ThingML) to Things' Machine Learning (ThingML2)
In this paper, we illustrate how to enhance an existing state-of-the-art modeling language and tool for the Internet of Things (IoT), called ThingML, to support machine learning on the modeling level.
ThingML+ Augmenting Model-Driven Software Engineering for the Internet of Things with Machine Learning
In this paper, we present the current position of the research project ML-Quadrat, which aims to extend the methodology, modeling language and tool support of ThingML - an open source modeling tool for IoT/CPS - to address Machine Learning needs for the IoT applications.
Efficient Robustness Certificates for Discrete Data: Sparsity-Aware Randomized Smoothing for Graphs, Images and More
Existing techniques for certifying the robustness of models for discrete data either work only for a small class of models or are general at the expense of efficiency or tightness.
Reachable Sets of Classifiers and Regression Models: (Non-)Robustness Analysis and Robust Training
This principle is highly versatile, as we show.
Deep Representation Learning and Clustering of Traffic Scenarios
Determining the traffic scenario space is a major challenge for the homologation and coverage assessment of automated driving functions.
Scaling Graph Neural Networks with Approximate PageRank
Graph neural networks (GNNs) have emerged as a powerful approach for solving many network mining tasks.
Scene Graph Reasoning for Visual Question Answering
We propose a novel method that approaches the task by performing context-driven, sequential reasoning based on the objects and their semantic and spatial relationships present in the scene.
Fast and Flexible Temporal Point Processes with Triangular Maps
Temporal point process (TPP) models combined with recurrent neural networks provide a powerful framework for modeling continuous-time event data.
Posterior Network: Uncertainty Estimation without OOD Samples via Density-Based Pseudo-Counts
The posterior distributions learned by PostNet accurately reflect uncertainty for in- and out-of-distribution data -- without requiring access to OOD data at training time.
Continual Learning with Bayesian Neural Networks for Non-Stationary Data
We represent the posterior approximation of the network weights by a diagonal Gaussian distribution and a complementary memory of raw data.
Graph Hawkes Neural Network for Forecasting on Temporal Knowledge Graphs
The Hawkes process has become a standard method for modeling self-exciting event sequences with different event types.
Directional Message Passing for Molecular Graphs
Each message is associated with a direction in coordinate space.
Ranked #2 on
Drug Discovery
on QM9
Oktoberfest Food Dataset
We release a realistic, diverse, and challenging dataset for object detection on images.
Uncertainty on Asynchronous Time Event Prediction
Asynchronous event sequences are the basis of many applications throughout different industries.
Certifiable Robustness to Graph Perturbations
Despite the exploding interest in graph neural networks there has been little effort to verify and improve their robustness.
Diffusion Improves Graph Learning
In this work, we remove the restriction of using only the direct neighbors by introducing a powerful, yet spatially localized graph convolution: Graph diffusion convolution (GDC).
Ranked #1 on
Node Classification
on AMZ Comp
Overlapping Community Detection with Graph Neural Networks
Community detection is a fundamental problem in machine learning.
Intensity-Free Learning of Temporal Point Processes
The standard way of learning in such models is by estimating the conditional intensity function.
Certifiable Robustness and Robust Training for Graph Convolutional Networks
Recent works show that Graph Neural Networks (GNNs) are highly non-robust with respect to adversarial attacks on both the graph structure and the node attributes, making their outcomes unreliable.
Ranked #18 on
Node Classification
on Pubmed
Adversarial Attacks on Node Embeddings
The goal of network representation learning is to learn low-dimensional node embeddings that capture the graph structure and are useful for solving downstream tasks.
Adversarial Attacks on Graph Neural Networks via Meta Learning
Deep learning models for graphs have advanced the state of the art on many tasks.
Pitfalls of Graph Neural Network Evaluation
We perform a thorough empirical evaluation of four prominent GNN models and show that considering different splits of the data leads to dramatically different rankings of models.
Multi-Source Neural Variational Inference
Learning from multiple sources of information is an important problem in machine-learning research.
Failing Loudly: An Empirical Study of Methods for Detecting Dataset Shift
We might hope that when faced with unexpected inputs, well-designed software systems would fire off warnings.
Predict then Propagate: Graph Neural Networks meet Personalized PageRank
We utilize this propagation procedure to construct a simple model, personalized propagation of neural predictions (PPNP), and its fast approximation, APPNP.
Ranked #1 on
Node Classification
on MS ACADEMIC
Mining Contrasting Quasi-Clique Patterns
Mining dense quasi-cliques is a well-known clustering task with applications ranging from social networks over collaboration graphs to document analysis.
Adversarial Attacks on Node Embeddings via Graph Poisoning
The goal of network representation learning is to learn low-dimensional node embeddings that capture the graph structure and are useful for solving downstream tasks.
Dual-Primal Graph Convolutional Networks
In recent years, there has been a surge of interest in developing deep learning methods for non-Euclidean structured data such as graphs.
Adversarial Attacks on Neural Networks for Graph Data
Even more, our attacks are transferable: the learned attacks generalize to other state-of-the-art node classification models and unsupervised approaches, and likewise are successful even when only limited knowledge about the graph is given.
NetGAN: Generating Graphs via Random Walks
NetGAN is able to produce graphs that exhibit well-known network patterns without explicitly specifying them in the model definition.
GraphGAN: Generating Graphs via Random Walks
Moreover, GraphGAN learns a semantic mapping from the latent input space to the generated graph's properties.
Introduction to Tensor Decompositions and their Applications in Machine Learning
Tensors are multidimensional arrays of numerical values and therefore generalize matrices to multiple dimensions.
Deep Gaussian Embedding of Graphs: Unsupervised Inductive Learning via Ranking
We propose Graph2Gauss - an approach that can efficiently learn versatile node embeddings on large scale (attributed) graphs that show strong performance on tasks such as link prediction and node classification.
Linearized and Single-Pass Belief Propagation
Often, we can answer such questions and label nodes in a network based on the labels of their neighbors and appropriate assumptions of homophily ("birds of a feather flock together") or heterophily ("opposites attract").
