Precision-recall divergence optimization for generative modeling with GANs and normalizing flows
Achieving a balance between image quality (precision) and diversity (recall) is a significant
challenge in the domain of generative models. Current state-of-the-art models primarily rely …
challenge in the domain of generative models. Current state-of-the-art models primarily rely …
Go with the flows: Mixtures of normalizing flows for point cloud generation and reconstruction
Recently Normalizing Flows (NFs) have demonstrated state-of-the-art performance on
modeling 3D point clouds while allowing sampling with arbitrary resolution at inference time …
modeling 3D point clouds while allowing sampling with arbitrary resolution at inference time …
Bayesian inversion, uncertainty analysis and interrogation using boosting variational inference
Geoscientists use observed data to estimate properties of the Earth's interior. This often
requires non‐linear inverse problems to be solved and uncertainties to be estimated …
requires non‐linear inverse problems to be solved and uncertainties to be estimated …
Synthetic Sensor Data Generation Exploiting Deep Learning Techniques and Multi-Modal Information
F Romanelli, F Martinelli - IEEE Sensors Letters, 2023 - ieeexplore.ieee.org
In recent years, deep learning techniques have revolutionized the field of data generation,
including the creation of synthetic sensor data. The ability to generate large quantities of …
including the creation of synthetic sensor data. The ability to generate large quantities of …
On the Quantification of Image Reconstruction Uncertainty without Training Data
Computational imaging plays a pivotal role in determining hidden information from sparse
measurements. A robust inverse solver is crucial to fully characterize the uncertainty induced …
measurements. A robust inverse solver is crucial to fully characterize the uncertainty induced …
Distributional gradient boosting machines
A März, T Kneib - arxiv preprint arxiv:2204.00778, 2022 - arxiv.org
We present a unified probabilistic gradient boosting framework for regression tasks that
models and predicts the entire conditional distribution of a univariate response variable as a …
models and predicts the entire conditional distribution of a univariate response variable as a …
Synthetic sensor measurement generation with noise learning and multi-modal information
F Romanelli, F Martinelli - IEEE Access, 2023 - ieeexplore.ieee.org
Deep learning has transformed data generation, particularly in creating synthetic sensor
data. This capability is invaluable in fields like autonomous driving, robotics, and computer …
data. This capability is invaluable in fields like autonomous driving, robotics, and computer …
On the Quantification of Image Reconstruction Uncertainty without Training Data
Computational imaging plays a pivotal role in determining hidden information from sparse
measurements. A robust inverse solver is crucial to fully characterize the uncertainty induced …
measurements. A robust inverse solver is crucial to fully characterize the uncertainty induced …
Training Algorithms for Mixtures of Normalizing Flows Check for updates
S Ciobanu - … Science and Network Engineering: Proceedings of …, 2023 - books.google.com
In this paper, we focus on how a probabilistic mixture of normalizing flows can be fitted. In
the literature, there are (at least) four approaches that do not necessarily provide an actual …
the literature, there are (at least) four approaches that do not necessarily provide an actual …
Training Algorithms for Mixtures of Normalizing Flows
S Ciobanu - International Conference on Data Science and Network …, 2023 - Springer
In this paper, we focus on how a probabilistic mixture of normalizing flows can be fitted. In
the literature, there are (at least) four approaches that do not necessarily provide an actual …
the literature, there are (at least) four approaches that do not necessarily provide an actual …