Search Results for author: Damian Borth

Found 50 papers, 23 papers with code

Sample Weight Estimation Using Meta-Updates for Online Continual Learning

1 code implementation29 Jan 2024 Hamed Hemati, Damian Borth

This is done by first estimating sample weight parameters for each sample in the mini-batch, then, updating the model with the adapted sample weights.

Continual Learning Meta-Learning

FedTabDiff: Federated Learning of Diffusion Probabilistic Models for Synthetic Mixed-Type Tabular Data Generation

1 code implementation11 Jan 2024 Timur Sattarov, Marco Schreyer, Damian Borth

Realistic synthetic tabular data generation encounters significant challenges in preserving privacy, especially when dealing with sensitive information in domains like finance and healthcare.

Attribute Denoising +1

Transformer-based Entity Legal Form Classification

1 code implementation19 Oct 2023 Alexander Arimond, Mauro Molteni, Dominik Jany, Zornitsa Manolova, Damian Borth, Andreas G. F. Hoepner

We propose the application of Transformer-based language models for classifying entity legal forms from raw legal entity names.

Data Integration text-classification +1

FinDiff: Diffusion Models for Financial Tabular Data Generation

1 code implementation4 Sep 2023 Timur Sattarov, Marco Schreyer, Damian Borth

The sharing of microdata, such as fund holdings and derivative instruments, by regulatory institutions presents a unique challenge due to strict data confidentiality and privacy regulations.

Fraud Detection Synthetic Data Generation

Ben-ge: Extending BigEarthNet with Geographical and Environmental Data

1 code implementation4 Jul 2023 Michael Mommert, Nicolas Kesseli, Joëlle Hanna, Linus Scheibenreif, Damian Borth, Begüm Demir

Based on this dataset, we showcase the value of combining different data modalities for the downstream tasks of patch-based land-use/land-cover classification and land-use/land-cover segmentation.

Earth Observation Land Cover Classification

Partial Hypernetworks for Continual Learning

1 code implementation19 Jun 2023 Hamed Hemati, Vincenzo Lomonaco, Davide Bacciu, Damian Borth

Inspired by latent replay methods in CL, we propose partial weight generation for the final layers of a model using hypernetworks while freezing the initial layers.

Continual Learning

Learning Emotional Representations from Imbalanced Speech Data for Speech Emotion Recognition and Emotional Text-to-Speech

no code implementations9 Jun 2023 Shijun Wang, Jón Guðnason, Damian Borth

Effective speech emotional representations play a key role in Speech Emotion Recognition (SER) and Emotional Text-To-Speech (TTS) tasks.

Speech Emotion Recognition

Sparsified Model Zoo Twins: Investigating Populations of Sparsified Neural Network Models

no code implementations26 Apr 2023 Dominik Honegger, Konstantin Schürholt, Damian Borth

With this paper, we address that gap by applying two popular sparsification methods on populations of models (so called model zoos) to create sparsified versions of the original zoos.

Class-Incremental Learning with Repetition

1 code implementation26 Jan 2023 Hamed Hemati, Andrea Cossu, Antonio Carta, Julio Hurtado, Lorenzo Pellegrini, Davide Bacciu, Vincenzo Lomonaco, Damian Borth

We propose two stochastic stream generators that produce a wide range of CIR streams starting from a single dataset and a few interpretable control parameters.

Class Incremental Learning Incremental Learning

Federated Continual Learning to Detect Accounting Anomalies in Financial Auditing

no code implementations26 Oct 2022 Marco Schreyer, Hamed Hemati, Damian Borth, Miklos A. Vasarhelyi

Our empirical results, using real-world datasets and combined federated continual learning strategies, demonstrate the learned model's ability to detect anomalies in audit settings of data distribution shifts.

Continual Learning

Hyper-Representations as Generative Models: Sampling Unseen Neural Network Weights

1 code implementation29 Sep 2022 Konstantin Schürholt, Boris Knyazev, Xavier Giró-i-Nieto, Damian Borth

Learning representations of neural network weights given a model zoo is an emerging and challenging area with many potential applications from model inspection, to neural architecture search or knowledge distillation.

Knowledge Distillation Neural Architecture Search +1

Model Zoos: A Dataset of Diverse Populations of Neural Network Models

1 code implementation29 Sep 2022 Konstantin Schürholt, Diyar Taskiran, Boris Knyazev, Xavier Giró-i-Nieto, Damian Borth

With this work, we publish a novel dataset of model zoos containing systematically generated and diverse populations of NN models for further research.

Classification Friction

Federated and Privacy-Preserving Learning of Accounting Data in Financial Statement Audits

no code implementations26 Aug 2022 Marco Schreyer, Timur Sattarov, Damian Borth

In this work, we propose a Federated Learning framework to train DL models on auditing relevant accounting data of multiple clients.

Federated Learning Privacy Preserving

Hyper-Representations for Pre-Training and Transfer Learning

1 code implementation22 Jul 2022 Konstantin Schürholt, Boris Knyazev, Xavier Giró-i-Nieto, Damian Borth

Learning representations of neural network weights given a model zoo is an emerging and challenging area with many potential applications from model inspection, to neural architecture search or knowledge distillation.

Knowledge Distillation Neural Architecture Search +4

Saliency Diversified Deep Ensemble for Robustness to Adversaries

no code implementations AAAI Workshop AdvML 2022 Alex Bogun, Dimche Kostadinov, Damian Borth

We empirically show a reduced transferability between ensemble members and improved performance compared to the state-of-the-art ensemble defense against medium and high strength white-box attacks.

Zero-shot Voice Conversion via Self-supervised Prosody Representation Learning

no code implementations27 Oct 2021 Shijun Wang, Dimche Kostadinov, Damian Borth

We then use the learned prosodic representations as conditional information to train and enhance our VC model for zero-shot conversion.

Disentanglement Voice Conversion

Multi-view Contrastive Self-Supervised Learning of Accounting Data Representations for Downstream Audit Tasks

no code implementations23 Sep 2021 Marco Schreyer, Timur Sattarov, Damian Borth

International audit standards require the direct assessment of a financial statement's underlying accounting transactions, referred to as journal entries.

Anomaly Detection Attribute +2

Heterogeneous Ensemble for ESG Ratings Prediction

no code implementations21 Sep 2021 Tim Krappel, Alex Bogun, Damian Borth

Such ratings allow them to make investment decisions in favor of sustainability.

Learning Interpretable Concept Groups in CNNs

1 code implementation21 Sep 2021 Saurabh Varshneya, Antoine Ledent, Robert A. Vandermeulen, Yunwen Lei, Matthias Enders, Damian Borth, Marius Kloft

We propose a novel training methodology -- Concept Group Learning (CGL) -- that encourages training of interpretable CNN filters by partitioning filters in each layer into concept groups, each of which is trained to learn a single visual concept.

Power Plant Classification from Remote Imaging with Deep Learning

no code implementations22 Jul 2021 Michael Mommert, Linus Scheibenreif, Joëlle Hanna, Damian Borth

Using a ResNet-50 deep learning model, we are able to achieve a mean accuracy of 90. 0% in distinguishing 10 different power plant types and a background class.

Classification

NoiseVC: Towards High Quality Zero-Shot Voice Conversion

no code implementations13 Apr 2021 Shijun Wang, Damian Borth

Voice conversion (VC) is a task that transforms voice from target audio to source without losing linguistic contents, it is challenging especially when source and target speakers are unseen during training (zero-shot VC).

Disentanglement Quantization +2

Continual Speaker Adaptation for Text-to-Speech Synthesis

no code implementations26 Mar 2021 Hamed Hemati, Damian Borth

The naive solution of sequential fine-tuning of a model for new speakers can lead to poor performance of older speakers.

Continual Learning Speech Synthesis +1

Leaking Sensitive Financial Accounting Data in Plain Sight using Deep Autoencoder Neural Networks

no code implementations13 Dec 2020 Marco Schreyer, Chistian Schulze, Damian Borth

Nowadays, organizations collect vast quantities of sensitive information in `Enterprise Resource Planning' (ERP) systems, such as accounting relevant transactions, customer master data, or strategic sales price information.

Characterization of Industrial Smoke Plumes from Remote Sensing Data

1 code implementation23 Nov 2020 Michael Mommert, Mario Sigel, Marcel Neuhausler, Linus Scheibenreif, Damian Borth

The major driver of global warming has been identified as the anthropogenic release of greenhouse gas (GHG) emissions from industrial activities.

Using IPA-Based Tacotron for Data Efficient Cross-Lingual Speaker Adaptation and Pronunciation Enhancement

no code implementations12 Nov 2020 Hamed Hemati, Damian Borth

Recent neural Text-to-Speech (TTS) models have been shown to perform very well when enough data is available.

Transfer Learning

Learning Sampling in Financial Statement Audits using Vector Quantised Autoencoder Neural Networks

no code implementations6 Aug 2020 Marco Schreyer, Timur Sattarov, Anita Gierbl, Bernd Reimer, Damian Borth

The audit of financial statements is designed to collect reasonable assurance that an issued statement is free from material misstatement 'true and fair presentation'.

Facial Recognition: A cross-national Survey on Public Acceptance, Privacy, and Discrimination

no code implementations15 Jul 2020 Léa Steinacker, Miriam Meckel, Genia Kostka, Damian Borth

With rapid advances in machine learning (ML), more of this technology is being deployed into the real world interacting with us and our environment.

An Investigation of the Weight Space to Monitor the Training Progress of Neural Networks

no code implementations18 Jun 2020 Konstantin Schürholt, Damian Borth

We show that the model trajectories can be separated and the order of checkpoints on the trajectories recovered, which may serve as a first step towards DNN model versioning.

Neural Networks and Value at Risk

no code implementations4 May 2020 Alexander Arimond, Damian Borth, Andreas Hoepner, Michael Klawunn, Stefan Weisheit

Utilizing a generative regime switching framework, we perform Monte-Carlo simulations of asset returns for Value at Risk threshold estimation.

Adversarial Learning of Deepfakes in Accounting

no code implementations9 Oct 2019 Marco Schreyer, Timur Sattarov, Bernd Reimer, Damian Borth

Second, we show that adversarial autoencoder neural networks are capable of learning a human interpretable model of journal entries that disentangles the entries latent generative factors.

Adversarial Attack

Detection of Accounting Anomalies in the Latent Space using Adversarial Autoencoder Neural Networks

4 code implementations2 Aug 2019 Marco Schreyer, Timur Sattarov, Christian Schulze, Bernd Reimer, Damian Borth

We demonstrate that such artificial neural networks are capable of learning a semantic meaningful representation of real-world journal entries.

Overcoming Missing and Incomplete Modalities with Generative Adversarial Networks for Building Footprint Segmentation

no code implementations9 Aug 2018 Benjamin Bischke, Patrick Helber, Florian König, Damian Borth, Andreas Dengel

This assumption limits the applications of multi-modal models since in practice the data collection process is likely to generate data with missing, incomplete or corrupted modalities.

Semantic Segmentation

What do Deep Networks Like to See?

1 code implementation CVPR 2018 Sebastian Palacio, Joachim Folz, Jörn Hees, Federico Raue, Damian Borth, Andreas Dengel

To do this, an autoencoder (AE) was fine-tuned on gradients from a pre-trained classifier with fixed parameters.

Image Classification

Multi-Task Learning for Segmentation of Building Footprints with Deep Neural Networks

1 code implementation18 Sep 2017 Benjamin Bischke, Patrick Helber, Joachim Folz, Damian Borth, Andreas Dengel

In this paper, we address the problem of preserving semantic segmentation boundaries in high resolution satellite imagery by introducing a new cascaded multi-task loss.

Multi-Task Learning Segmentation +1

Detection of Anomalies in Large Scale Accounting Data using Deep Autoencoder Networks

4 code implementations15 Sep 2017 Marco Schreyer, Timur Sattarov, Damian Borth, Andreas Dengel, Bernd Reimer

Learning to detect fraud in large-scale accounting data is one of the long-standing challenges in financial statement audits or fraud investigations.

Attribute

EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classification

8 code implementations31 Aug 2017 Patrick Helber, Benjamin Bischke, Andreas Dengel, Damian Borth

We present a novel dataset based on Sentinel-2 satellite images covering 13 spectral bands and consisting out of 10 classes with in total 27, 000 labeled and geo-referenced images.

Earth Observation General Classification +1

AudioPairBank: Towards A Large-Scale Tag-Pair-Based Audio Content Analysis

no code implementations13 Jul 2016 Sebastian Sager, Benjamin Elizalde, Damian Borth, Christian Schulze, Bhiksha Raj, Ian Lane

One contribution is the previously unavailable documentation of the challenges and implications of collecting audio recordings with these type of labels.

TAG

Mapping Images to Sentiment Adjective Noun Pairs with Factorized Neural Nets

no code implementations21 Nov 2015 Takuya Narihira, Damian Borth, Stella X. Yu, Karl Ni, Trevor Darrell

We consider the visual sentiment task of mapping an image to an adjective noun pair (ANP) such as "cute baby".

Image Captioning

The YLI-MED Corpus: Characteristics, Procedures, and Plans

no code implementations13 Mar 2015 Julia Bernd, Damian Borth, Benjamin Elizalde, Gerald Friedland, Heather Gallagher, Luke Gottlieb, Adam Janin, Sara Karabashlieva, Jocelyn Takahashi, Jennifer Won

The YLI Multimedia Event Detection corpus is a public-domain index of videos with annotations and computed features, specialized for research in multimedia event detection (MED), i. e., automatically identifying what's happening in a video by analyzing the audio and visual content.

Descriptive Event Detection

YFCC100M: The New Data in Multimedia Research

2 code implementations5 Mar 2015 Bart Thomee, David A. Shamma, Gerald Friedland, Benjamin Elizalde, Karl Ni, Douglas Poland, Damian Borth, Li-Jia Li

We present the Yahoo Flickr Creative Commons 100 Million Dataset (YFCC100M), the largest public multimedia collection that has ever been released.

Multimedia Computers and Society H.3.7

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