Search Results for author: Gerhard Wunder

Found 12 papers, 3 papers with code

On Gradient-like Explanation under a Black-box Setting: When Black-box Explanations Become as Good as White-box

1 code implementation18 Aug 2023 Yi Cai, Gerhard Wunder

Attribution methods shed light on the explainability of data-driven approaches such as deep learning models by uncovering the most influential features in a to-be-explained decision.

Differentially Private Synthetic Data Generation via Lipschitz-Regularised Variational Autoencoders

no code implementations22 Apr 2023 Benedikt Groß, Gerhard Wunder

It is often overlooked though that generative models are prone to memorising many details of individual training records and often generate synthetic data that too closely resembles the underlying sensitive training data, hence violating strong privacy regulations as, e. g., encountered in health care.

Privacy Preserving Synthetic Data Generation

Explaining text classifiers through progressive neighborhood approximation with realistic samples

no code implementations11 Feb 2023 Yi Cai, Arthur Zimek, Eirini Ntoutsi, Gerhard Wunder

The importance of neighborhood construction in local explanation methods has been already highlighted in the literature.

Power of Explanations: Towards automatic debiasing in hate speech detection

1 code implementation7 Sep 2022 Yi Cai, Arthur Zimek, Gerhard Wunder, Eirini Ntoutsi

Hate speech detection is a common downstream application of natural language processing (NLP) in the real world.

Fairness Hate Speech Detection

An Algorithm for Exact Numerical Age-of-Information Evaluation in Multi-Agent Systems

no code implementations11 Jun 2022 Richard Schoeffauer, Gerhard Wunder

We present an algorithm for the numerical evaluation of the state-space distribution of an Age-of-Information network.

A Reverse Jensen Inequality Result with Application to Mutual Information Estimation

no code implementations12 Nov 2021 Gerhard Wunder, Benedikt Groß, Rick Fritschek, Rafael F. Schaefer

The Jensen inequality is a widely used tool in a multitude of fields, such as for example information theory and machine learning.

Mutual Information Estimation

Explicit CSI Feedback Compression via Learned Approximate Message Passing

no code implementations12 Oct 2021 Benedikt Groß, Rana Ahmed Salem, Thorsten Wild, Gerhard Wunder

Explicit channel state information at the transmitter side is helpful to improve downlink precoding performance for multi-user MIMO systems.

Reinforce Security: A Model-Free Approach Towards Secure Wiretap Coding

no code implementations1 Jun 2021 Rick Fritschek, Rafael F. Schaefer, Gerhard Wunder

The use of deep learning-based techniques for approximating secure encoding functions has attracted considerable interest in wireless communications due to impressive results obtained for general coding and decoding tasks for wireless communication systems.

Mutual Information Estimation reinforcement-learning +1

ComPass: Proximity Aware Common Passphrase Agreement Protocol for Wi-Fi devices Using Physical Layer Security

no code implementations11 Mar 2021 Khan Reaz, Gerhard Wunder

Secure and scalable device provisioning is a notorious challenge in Wi-Fi.

Cryptography and Security

Using AoI Forecasts in Communicating and Robust Distributed Model-Predictive Control

no code implementations9 Mar 2021 Jannik Hahn, Richard Schoeffauer, Gerhard Wunder, Olaf Stursberg

By contrast, the present approachestablishes a robust distributed model-predictive controlscheme, in which the local subsystem controllers oper-ate under the assumption of a variable communicationschedule that is predicted by a network controller.

Systems and Control Systems and Control

Deep Learning for Channel Coding via Neural Mutual Information Estimation

1 code implementation7 Mar 2019 Rick Fritschek, Rafael F. Schaefer, Gerhard Wunder

However, one of the drawbacks of current learning approaches is that a differentiable channel model is needed for the training of the underlying neural networks.

Decoder Mutual Information Estimation

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