34 citations · 83 across the 11 of their papers we have counts for
14 papers
Three Conjectures on Binary Channels for the Doubly Symmetric Binary Source
Georg Pichler
We settle three conjectures concerning a doubly symmetric binary source with crossover . Consider Markov chains with binary, and let …
A Data-Driven Measure of Relative Uncertainty for Misclassification Detection
Eduardo Dadalto, Marco Romanelli, Georg Pichler +1
Misclassification detection is an important problem in machine learning, as it allows for the identification of instances where the model's predictions are unreliable. However, con…
Leveraging Adversarial Examples to Quantify Membership Information Leakage
Ganesh Del Grosso, Hamid Jalalzai, Georg Pichler +2
The use of personal data for training machine learning systems comes with a privacy threat and measuring the level of privacy of a model is one of the major challenges in machine l…
Perfectly Accurate Membership Inference by a Dishonest Central Server in Federated Learning
Georg Pichler, Marco Romanelli, Leonardo Rey Vega +1
Federated Learning is expected to provide strong privacy guarantees, as only gradients or model parameters but no plain text training data is ever exchanged either between the clie…
A Differential Entropy Estimator for Training Neural Networks
Georg Pichler, Pierre Colombo, Malik Boudiaf +2
Mutual Information (MI) has been widely used as a loss regularizer for training neural networks. This has been particularly effective when learn disentangled or compressed represen…
Modelling the Utility of Group Testing for Public Health Surveillance
Günther Koliander, Georg Pichler
In epidemic or pandemic situations, resources for testing the infection status of individuals may be scarce. Although group testing can help to significantly increase testing capab…
Bounding Information Leakage in Machine Learning
Ganesh Del Grosso, Georg Pichler, Catuscia Palamidessi +1
Recently, it has been shown that Machine Learning models can leak sensitive information about their training data. This information leakage is exposed through membership and attrib…
Privacy-Preserving Synthetic Smart Meters Data
Ganesh Del Grosso, Georg Pichler, Pablo Piantanida
Power consumption data is very useful as it allows to optimize power grids, detect anomalies and prevent failures, on top of being useful for diverse research purposes. However, th…
On the Estimation of Information Measures of Continuous Distributions
Georg Pichler, Pablo Piantanida, Günther Koliander
The estimation of information measures of continuous distributions based on samples is a fundamental problem in statistics and machine learning. In this paper, we analyze estimates…
On Direct Distribution Matching for Adapting Segmentation Networks
Georg Pichler, Jose Dolz, Ismail Ben Ayed +1
Minimization of distribution matching losses is a principled approach to domain adaptation in the context of image classification. However, it is largely overlooked in adapting seg…
Information Bottleneck on General Alphabets
Georg Pichler, Günther Koliander
We prove rigorously a source coding theorem that can probably be considered folklore, a generalization to arbitrary alphabets of a problem motivated by the Information Bottleneck m…
Dictator Functions Maximize Mutual Information
Georg Pichler, Pablo Piantanida, Gerald Matz
Let denote independent, identically distributed copies of two arbitrarily correlated Rademacher random variables . We prove that the inequality…
Distributed Information-Theoretic Clustering
Georg Pichler, Pablo Piantanida, Gerald Matz
We study a novel multi-terminal source coding setup motivated by the biclustering problem. Two separate encoders observe two i.i.d. sequences and , respectively. The goa…
Entropy and Source Coding for Integer-Dimensional Singular Random Variables
Günther Koliander, Georg Pichler, Erwin Riegler +1
Entropy and differential entropy are important quantities in information theory. A tractable extension to singular random variables-which are neither discrete nor continuous-has no…