Search Results for author: Asja Fischer

Found 53 papers, 23 papers with code

Detecting Compositionally Out-of-Distribution Examples in Semantic Parsing

no code implementations Findings (EMNLP) 2021 Denis Lukovnikov, Sina Daubener, Asja Fischer

While neural networks are ubiquitous in state-of-the-art semantic parsers, it has been shown that most standard models suffer from dramatic performance losses when faced with compositionally out-of-distribution (OOD) data.

Out of Distribution (OOD) Detection Semantic Parsing

Layout-to-Image Generation with Localized Descriptions using ControlNet with Cross-Attention Control

no code implementations20 Feb 2024 Denis Lukovnikov, Asja Fischer

While text-to-image diffusion models can generate highquality images from textual descriptions, they generally lack fine-grained control over the visual composition of the generated images.

Layout-to-Image Generation

AEROBLADE: Training-Free Detection of Latent Diffusion Images Using Autoencoder Reconstruction Error

1 code implementation31 Jan 2024 Jonas Ricker, Denis Lukovnikov, Asja Fischer

A key enabler for generating high-resolution images with low computational cost has been the development of latent diffusion models (LDMs).

Denoising

Benchmarking the Fairness of Image Upsampling Methods

no code implementations24 Jan 2024 Mike Laszkiewicz, Imant Daunhawer, Julia E. Vogt, Asja Fischer, Johannes Lederer

Recent years have witnessed a rapid development of deep generative models for creating synthetic media, such as images and videos.

Benchmarking Fairness

A Representative Study on Human Detection of Artificially Generated Media Across Countries

1 code implementation10 Dec 2023 Joel Frank, Franziska Herbert, Jonas Ricker, Lea Schönherr, Thorsten Eisenhofer, Asja Fischer, Markus Dürmuth, Thorsten Holz

To further understand which factors influence people's ability to detect generated media, we include personal variables, chosen based on a literature review in the domains of deepfake and fake news research.

Face Swapping Human Detection

Learning Sparse Codes with Entropy-Based ELBOs

no code implementations3 Nov 2023 Dmytro Velychko, Simon Damm, Asja Fischer, Jörg Lücke

Our main contributions are theoretical, however, and they are twofold: (1) for non-trivial posterior approximations, we provide the (to the knowledge of the authors) first analytical ELBO objective for standard probabilistic sparse coding; and (2) we provide the first demonstration on how a recently shown convergence of the ELBO to entropy sums can be used for learning.

Uncertainty-weighted Loss Functions for Improved Adversarial Attacks on Semantic Segmentation

1 code implementation26 Oct 2023 Kira Maag, Asja Fischer

State-of-the-art deep neural networks have been shown to be extremely powerful in a variety of perceptual tasks like semantic segmentation.

Image Segmentation Segmentation +1

Layer-wise Linear Mode Connectivity

1 code implementation13 Jul 2023 Linara Adilova, Maksym Andriushchenko, Michael Kamp, Asja Fischer, Martin Jaggi

Averaging neural network parameters is an intuitive method for fusing the knowledge of two independent models.

Federated Learning Linear Mode Connectivity

Set-Membership Inference Attacks using Data Watermarking

no code implementations22 Jun 2023 Mike Laszkiewicz, Denis Lukovnikov, Johannes Lederer, Asja Fischer

In this work, we propose a set-membership inference attack for generative models using deep image watermarking techniques.

Inference Attack Membership Inference Attack

DistriBlock: Identifying adversarial audio samples by leveraging characteristics of the output distribution

no code implementations26 May 2023 Matías Pizarro, Dorothea Kolossa, Asja Fischer

Adversarial attacks can mislead automatic speech recognition (ASR) systems into predicting an arbitrary target text, thus posing a clear security threat.

Automatic Speech Recognition Automatic Speech Recognition (ASR) +1

Single-Model Attribution of Generative Models Through Final-Layer Inversion

no code implementations26 May 2023 Mike Laszkiewicz, Jonas Ricker, Johannes Lederer, Asja Fischer

Recent breakthroughs in generative modeling have sparked interest in practical single-model attribution.

Anomaly Detection

Uncertainty-based Detection of Adversarial Attacks in Semantic Segmentation

1 code implementation22 May 2023 Kira Maag, Asja Fischer

State-of-the-art deep neural networks have proven to be highly powerful in a broad range of tasks, including semantic image segmentation.

Image Classification Image Segmentation +2

Towards the Detection of Diffusion Model Deepfakes

1 code implementation26 Oct 2022 Jonas Ricker, Simon Damm, Thorsten Holz, Asja Fischer

However, relatively little attention has been paid to the detection of DM-generated images, which is critical to prevent adverse impacts on our society.

Attribute Image Generation

Marginal Tail-Adaptive Normalizing Flows

1 code implementation21 Jun 2022 Mike Laszkiewicz, Johannes Lederer, Asja Fischer

Learning the tail behavior of a distribution is a notoriously difficult problem.

How Sampling Impacts the Robustness of Stochastic Neural Networks

no code implementations22 Apr 2022 Sina Däubener, Asja Fischer

Stochastic neural networks (SNNs) are random functions whose predictions are gained by averaging over multiple realizations.

Adversarial Attack

Copula-Based Normalizing Flows

1 code implementation ICML Workshop INNF 2021 Mike Laszkiewicz, Johannes Lederer, Asja Fischer

Normalizing flows, which learn a distribution by transforming the data to samples from a Gaussian base distribution, have proven powerful density approximations.

Gated Relational Graph Attention Networks

no code implementations1 Jan 2021 Denis Lukovnikov, Asja Fischer

Relational Graph Neural Networks (GNN) are a class of GNN that are capable of handling multi-relational graphs.

Graph Attention Long-range modeling

SmoothLRP: Smoothing Explanations of Neural Network Decisions by Averaging over Stochastic Input Variations

no code implementations1 Jan 2021 Arne Peter Raulf, Ben Luis Hack, Sina Däubener, Axel Mosig, Asja Fischer

With the excessive use of neural networks in safety critical domains the need for understandable explanations of their predictions is rising.

Wasserstein Dropout

1 code implementation23 Dec 2020 Joachim Sicking, Maram Akila, Maximilian Pintz, Tim Wirtz, Asja Fischer, Stefan Wrobel

Despite of its importance for safe machine learning, uncertainty quantification for neural networks is far from being solved.

Object Detection regression +1

Efficient Calculation of Adversarial Examples for Bayesian Neural Networks

no code implementations pproximateinference AABI Symposium 2021 Sina Däubener, Joel Frank, Thorsten Holz, Asja Fischer

In this paper we propose to efficiently attack Bayesian neural networks with adversarial examples calculated for a deterministic network with parameters given by the mean of the posterior distribution.

The ELBO of Variational Autoencoders Converges to a Sum of Three Entropies

1 code implementation28 Oct 2020 Simon Damm, Dennis Forster, Dmytro Velychko, Zhenwen Dai, Asja Fischer, Jörg Lücke

Here we show that for standard (i. e., Gaussian) VAEs the ELBO converges to a value given by the sum of three entropies: the (negative) entropy of the prior distribution, the expected (negative) entropy of the observable distribution, and the average entropy of the variational distributions (the latter is already part of the ELBO).

Improving the Long-Range Performance of Gated Graph Neural Networks

no code implementations19 Jul 2020 Denis Lukovnikov, Jens Lehmann, Asja Fischer

Many popular variants of graph neural networks (GNNs) that are capable of handling multi-relational graphs may suffer from vanishing gradients.

Characteristics of Monte Carlo Dropout in Wide Neural Networks

no code implementations10 Jul 2020 Joachim Sicking, Maram Akila, Tim Wirtz, Sebastian Houben, Asja Fischer

Monte Carlo (MC) dropout is one of the state-of-the-art approaches for uncertainty estimation in neural networks (NNs).

Bayesian Inference Gaussian Processes

On the convergence of the Metropolis algorithm with fixed-order updates for multivariate binary probability distributions

no code implementations26 Jun 2020 Kai Brügge, Asja Fischer, Christian Igel

We propose a modified Metropolis transition operator that behaves almost always identically to the standard Metropolis operator and prove that it ensures irreducibility and convergence to the limiting distribution in the multivariate binary case with fixed-order updates.

Detecting Adversarial Examples for Speech Recognition via Uncertainty Quantification

1 code implementation24 May 2020 Sina Däubener, Lea Schönherr, Asja Fischer, Dorothea Kolossa

The neural networks for uncertainty quantification simultaneously diminish the vulnerability to the attack, which is reflected in a lower recognition accuracy of the malicious target text in comparison to a standard hybrid ASR system.

Automatic Speech Recognition Automatic Speech Recognition (ASR) +3

Unsupervised Cross-Domain Speech-to-Speech Conversion with Time-Frequency Consistency

no code implementations15 May 2020 Mohammad Asif Khan, Fabien Cardinaux, Stefan Uhlich, Marc Ferras, Asja Fischer

This procedure bears the problem that the generated magnitude spectrogram may not be consistent, which is required for finding a phase such that the full spectrogram has a natural-sounding speech waveform.

Generative Adversarial Network

Thresholded Adaptive Validation: Tuning the Graphical Lasso for Graph Recovery

1 code implementation1 May 2020 Mike Laszkiewicz, Asja Fischer, Johannes Lederer

Many Machine Learning algorithms are formulated as regularized optimization problems, but their performance hinges on a regularization parameter that needs to be calibrated to each application at hand.

Leveraging Frequency Analysis for Deep Fake Image Recognition

1 code implementation ICML 2020 Joel Frank, Thorsten Eisenhofer, Lea Schönherr, Asja Fischer, Dorothea Kolossa, Thorsten Holz

Based on this analysis, we demonstrate how the frequency representation can be used to identify deep fake images in an automated way, surpassing state-of-the-art methods.

Image Forensics

End-to-End Entity Linking and Disambiguation leveraging Word and Knowledge Graph Embeddings

no code implementations25 Feb 2020 Rostislav Nedelchev, Debanjan Chaudhuri, Jens Lehmann, Asja Fischer

Entity linking - connecting entity mentions in a natural language utterance to knowledge graph (KG) entities is a crucial step for question answering over KGs.

Entity Disambiguation Entity Linking +5

Introduction to Neural Network based Approaches for Question Answering over Knowledge Graphs

no code implementations22 Jul 2019 Nilesh Chakraborty, Denis Lukovnikov, Gaurav Maheshwari, Priyansh Trivedi, Jens Lehmann, Asja Fischer

Question answering has emerged as an intuitive way of querying structured data sources, and has attracted significant advancements over the years.

Knowledge Graphs Question Answering

Compound Density Networks

no code implementations ICLR 2019 Agustinus Kristiadi, Asja Fischer

Despite the huge success of deep neural networks (NNs), finding good mechanisms for quantifying their prediction uncertainty is still an open problem.

Predictive Uncertainty Quantification with Compound Density Networks

no code implementations4 Feb 2019 Agustinus Kristiadi, Sina Däubener, Asja Fischer

Despite the huge success of deep neural networks (NNs), finding good mechanisms for quantifying their prediction uncertainty is still an open problem.

Bayesian Inference Uncertainty Quantification

Translating Natural Language to SQL using Pointer-Generator Networks and How Decoding Order Matters

no code implementations13 Nov 2018 Denis Lukovnikov, Nilesh Chakraborty, Jens Lehmann, Asja Fischer

Translating natural language to SQL queries for table-based question answering is a challenging problem and has received significant attention from the research community.

Question Answering Semantic Parsing

Learning to Rank Query Graphs for Complex Question Answering over Knowledge Graphs

1 code implementation2 Nov 2018 Gaurav Maheshwari, Priyansh Trivedi, Denis Lukovnikov, Nilesh Chakraborty, Asja Fischer, Jens Lehmann

In this paper, we conduct an empirical investigation of neural query graph ranking approaches for the task of complex question answering over knowledge graphs.

Graph Ranking Knowledge Graphs +3

On the Relation Between the Sharpest Directions of DNN Loss and the SGD Step Length

1 code implementation ICLR 2019 Stanisław Jastrzębski, Zachary Kenton, Nicolas Ballas, Asja Fischer, Yoshua Bengio, Amos Storkey

When studying the SGD dynamics in relation to the sharpest directions in this initial phase, we find that the SGD step is large compared to the curvature and commonly fails to minimize the loss along the sharpest directions.

Relation

Incorporating Literals into Knowledge Graph Embeddings

1 code implementation3 Feb 2018 Agustinus Kristiadi, Mohammad Asif Khan, Denis Lukovnikov, Jens Lehmann, Asja Fischer

Most of the existing work on embedding (or latent feature) based knowledge graph analysis focuses mainly on the relations between entities.

Entity Embeddings Knowledge Graph Embeddings +2

Three Factors Influencing Minima in SGD

no code implementations ICLR 2018 Stanisław Jastrzębski, Zachary Kenton, Devansh Arpit, Nicolas Ballas, Asja Fischer, Yoshua Bengio, Amos Storkey

In particular we find that the ratio of learning rate to batch size is a key determinant of SGD dynamics and of the width of the final minima, and that higher values of the ratio lead to wider minima and often better generalization.

Memorization Open-Ended Question Answering

On the regularization of Wasserstein GANs

2 code implementations ICLR 2018 Henning Petzka, Asja Fischer, Denis Lukovnicov

Since their invention, generative adversarial networks (GANs) have become a popular approach for learning to model a distribution of real (unlabeled) data.

Graph-based Predictable Feature Analysis

no code implementations1 Feb 2016 Björn Weghenkel, Asja Fischer, Laurenz Wiskott

We propose graph-based predictable feature analysis (GPFA), a new method for unsupervised learning of predictable features from high-dimensional time series, where high predictability is understood very generically as low variance in the distribution of the next data point given the previous ones.

Graph Embedding Time Series +1

Early Inference in Energy-Based Models Approximates Back-Propagation

no code implementations9 Oct 2015 Yoshua Bengio, Asja Fischer

We show that Langevin MCMC inference in an energy-based model with latent variables has the property that the early steps of inference, starting from a stationary point, correspond to propagating error gradients into internal layers, similarly to back-propagation.

Population-Contrastive-Divergence: Does Consistency help with RBM training?

no code implementations6 Oct 2015 Oswin Krause, Asja Fischer, Christian Igel

Compared to CD, it leads to a consistent estimate and may have a significantly lower bias.

STDP as presynaptic activity times rate of change of postsynaptic activity

no code implementations19 Sep 2015 Yoshua Bengio, Thomas Mesnard, Asja Fischer, Saizheng Zhang, Yuhuai Wu

We introduce a weight update formula that is expressed only in terms of firing rates and their derivatives and that results in changes consistent with those associated with spike-timing dependent plasticity (STDP) rules and biological observations, even though the explicit timing of spikes is not needed.

Bidirectional Helmholtz Machines

1 code implementation12 Jun 2015 Jorg Bornschein, Samira Shabanian, Asja Fischer, Yoshua Bengio

We present a lower-bound for the likelihood of this model and we show that optimizing this bound regularizes the model so that the Bhattacharyya distance between the bottom-up and top-down approximate distributions is minimized.

Difference Target Propagation

1 code implementation23 Dec 2014 Dong-Hyun Lee, Saizheng Zhang, Asja Fischer, Yoshua Bengio

Back-propagation has been the workhorse of recent successes of deep learning but it relies on infinitesimal effects (partial derivatives) in order to perform credit assignment.

How to Center Binary Deep Boltzmann Machines

1 code implementation6 Nov 2013 Jan Melchior, Asja Fischer, Laurenz Wiskott

This work analyzes centered binary Restricted Boltzmann Machines (RBMs) and binary Deep Boltzmann Machines (DBMs), where centering is done by subtracting offset values from visible and hidden variables.

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