Probabilistic Circuits That Know What They Don't Know

2 code implementations • 13 Feb 2023 • Fabrizio Ventola, Steven Braun, Zhongjie Yu, Martin Mundt, Kristian Kersting

In contrast to neural networks, they are often assumed to be well-calibrated and robust to out-of-distribution (OOD) data.

Uncertainty Quantification

DAFNe: A One-Stage Anchor-Free Approach for Oriented Object Detection

1 code implementation • 13 Sep 2021 • Steven Lang, Fabrizio Ventola, Kristian Kersting

We present DAFNe, a Dense one-stage Anchor-Free deep Network for oriented object detection.

object-detection Object Detection +3

RECOWNs: Probabilistic Circuits for Trustworthy Time Series Forecasting

no code implementations • 8 Jun 2021 • Nils Thoma, Zhongjie Yu, Fabrizio Ventola, Kristian Kersting

Time series forecasting is a relevant task that is performed in several real-world scenarios such as product sales analysis and prediction of energy demand.

Time Series Time Series Forecasting

User-Level Label Leakage from Gradients in Federated Learning

2 code implementations • 19 May 2021 • Aidmar Wainakh, Fabrizio Ventola, Till Müßig, Jens Keim, Carlos Garcia Cordero, Ephraim Zimmer, Tim Grube, Kristian Kersting, Max Mühlhäuser

Specifically, we investigate Label Leakage from Gradients (LLG), a novel attack to extract the labels of the users' training data from their shared gradients.

Federated Learning

Random Sum-Product Forests with Residual Links

no code implementations • 8 Aug 2019 • Fabrizio Ventola, Karl Stelzner, Alejandro Molina, Kristian Kersting

Tractable yet expressive density estimators are a key building block of probabilistic machine learning.