arXiv subjects

Philipp Koch

Publications and source records attributed to Philipp Koch.

At least 19 recordsLinked to original sources

arXiv2025

Augmenting the availability of historical GDP per capita estimates through machine learning

Can we use data on the biographies of historical figures to estimate the GDP per capita of countries and regions? Here we introduce a machine learning method to estimate the GDP per capita of dozens of countries and hundreds of regions in Europe and North America for the past 700 years starting from data on the places of birth, death, and occupations of hundreds of thousands of historical figures. We build an elastic net regression model to perform feature selection and generate out-of-sample estimates that explain 90% of the variance in known historical income levels. We use this model to generate GDP per capita estimates for countries, regions, and time periods for which this data is not available and externally validate our estimates by comparing them with four proxies of economic output: urbanization rates in the past 500 years, body height in the 18th century, wellbeing in 1850, and church building activity in the 14th and 15th century. Additionally, we show our estimates reproduce the well-known reversal of fortune between southwestern and northwestern Europe between 1300 and 1800 and find this is largely driven by countries and regions engaged in Atlantic trade. These findings validate the use of fine-grained biographical data as a method to produce historical GDP per capita estimates. We publish our estimates with confidence intervals together with all collected source data in a comprehensive dataset.

econ.GN↗

arXiv2023

Estimating Digital Product Trade through Corporate Revenue Data

Despite global efforts to harmonize international trade statistics, our understanding of digital trade and its implications remains limited. Here, we introduce a method to estimate bilateral exports and imports for dozens of sectors starting from the corporate revenue data of large digital firms. This method allows us to provide estimates for digitally ordered and delivered trade involving digital goods (e.g. video games), productized services (e.g. digital advertising), and digital intermediation fees (e.g. hotel rental), which together we call digital products. We use these estimates to study five key aspects of digital trade. We find that, compared to trade in physical goods, digital product exports are more spatially concentrated, have been growing faster, and can offset trade balance estimates, like the United States trade deficit on physical goods. We also find that countries that have decoupled economic growth from greenhouse gas emissions tend to have larger digital exports and that digital products exports contribute positively to the complexity of economies. This method, dataset, and findings provide a new lens to understand the impact of international trade in digital products.

econ.GN↗

arXiv2023

A tailored Handwritten-Text-Recognition System for Medieval Latin

The Bavarian Academy of Sciences and Humanities aims to digitize its Medieval Latin Dictionary. This dictionary entails record cards referring to lemmas in medieval Latin, a low-resource language. A crucial step of the digitization process is the Handwritten Text Recognition (HTR) of the handwritten lemmas found on these record cards. In our work, we introduce an end-to-end pipeline, tailored to the medieval Latin dictionary, for locating, extracting, and transcribing the lemmas. We employ two state-of-the-art (SOTA) image segmentation models to prepare the initial data set for the HTR task. Furthermore, we experiment with different transformer-based models and conduct a set of experiments to explore the capabilities of different combinations of vision encoders with a GPT-2 decoder. Additionally, we also apply extensive data augmentation resulting in a highly competitive model. The best-performing setup achieved a Character Error Rate (CER) of 0.015, which is even superior to the commercial Google Cloud Vision model, and shows more stable performance.

cs.CV↗

arXiv2023

Multimodal Deep Learning

This book is the result of a seminar in which we reviewed multimodal approaches and attempted to create a solid overview of the field, starting with the current state-of-the-art approaches in the two subfields of Deep Learning individually. Further, modeling frameworks are discussed where one modality is transformed into the other, as well as models in which one modality is utilized to enhance representation learning for the other. To conclude the second part, architectures with a focus on handling both modalities simultaneously are introduced. Finally, we also cover other modalities as well as general-purpose multi-modal models, which are able to handle different tasks on different modalities within one unified architecture. One interesting application (Generative Art) eventually caps off this booklet.

cs.CL↗

arXiv2023

L-SeqSleepNet: Whole-cycle Long Sequence Modelling for Automatic Sleep Staging

Human sleep is cyclical with a period of approximately 90 minutes, implying long temporal dependency in the sleep data. Yet, exploring this long-term dependency when developing sleep staging models has remained untouched. In this work, we show that while encoding the logic of a whole sleep cycle is crucial to improve sleep staging performance, the sequential modelling approach in existing state-of-the-art deep learning models are inefficient for that purpose. We thus introduce a method for efficient long sequence modelling and propose a new deep learning model, L-SeqSleepNet, which takes into account whole-cycle sleep information for sleep staging. Evaluating L-SeqSleepNet on four distinct databases of various sizes, we demonstrate state-of-the-art performance obtained by the model over three different EEG setups, including scalp EEG in conventional Polysomnography (PSG), in-ear EEG, and around-the-ear EEG (cEEGrid), even with a single EEG channel input. Our analyses also show that L-SeqSleepNet is able to alleviate the predominance of N2 sleep (the major class in terms of classification) to bring down errors in other sleep stages. Moreover the network becomes much more robust, meaning that for all subjects where the baseline method had exceptionally poor performance, their performance are improved significantly. Finally, the computation time only grows at a sub-linear rate when the sequence length increases.

eess.SP↗

arXiv2022

The Role of Immigrants, Emigrants, and Locals in the Historical Formation of European Knowledge Agglomerations

Did migrants make Paris a Mecca for the arts and Vienna a beacon of classical music? Or was their rise a pure consequence of local actors? Here, we use data on more than 22,000 historical individuals born between the years 1000 and 2000 to estimate the contribution of famous immigrants, emigrants, and locals to the knowledge specializations of European regions. We find that the probability that a region develops or keeps specialization in an activity (based on the birth of famous physicists, painters, etc.) grows with both, the presence of immigrants with knowledge on that activity and immigrants with knowledge in related activities. In contrast, we do not find robust evidence that the presence of locals with related knowledge explains entries and/or exits. We address some endogeneity concerns using fixed-effects models considering any location-period-activity specific factors (e.g. the presence of a new university attracting scientists).

econ.GN↗

arXiv2022

Multidimensional Economic Complexity and Inclusive Green Growth

To achieve inclusive green growth, countries need to consider a multiplicity of economic, social, and environmental factors. These are often captured by metrics of economic complexity derived from the geography of trade, thus missing key information on innovative activities. To bridge this gap, we combine trade data with data on patent applications and research publications to build models that significantly and robustly improve the ability of economic complexity metrics to explain international variations in inclusive green growth. We show that measures of complexity built on trade and patent data combine to explain future economic growth and income inequality and that countries that score high in all three metrics tend to exhibit lower emission intensities. These findings illustrate how the geography of trade, technology, and research combine to explain inclusive green growth.

econ.GN↗

arXiv2022

Polyphonic audio event detection: multi-label or multi-class multi-task classification problem?

Polyphonic events are the main error source of audio event detection (AED) systems. In deep-learning context, the most common approach to deal with event overlaps is to treat the AED task as a multi-label classification problem. By doing this, we inherently consider multiple one-vs.-rest classification problems, which are jointly solved by a single (i.e. shared) network. In this work, to better handle polyphonic mixtures, we propose to frame the task as a multi-class classification problem by considering each possible label combination as one class. To circumvent the large number of arising classes due to combinatorial explosion, we divide the event categories into multiple groups and construct a multi-task problem in a divide-and-conquer fashion, where each of the tasks is a multi-class classification problem. A network architecture is then devised for multi-class multi-task modelling. The network is composed of a backbone subnet and multiple task-specific subnets. The task-specific subnets are designed to learn time-frequency and channel attention masks to extract features for the task at hand from the common feature maps learned by the backbone. Experiments on the TUT-SED-Synthetic-2016 with high degree of event overlap show that the proposed approach results in more favorable performance than the common multi-label approach.

eess.AS↗

arXiv2021

Rapidly-Exploring Random Graph Next-Best View Exploration for Ground Vehicles

In this paper, a novel approach is introduced which utilizes a Rapidly-exploring Random Graph to improve sampling-based autonomous exploration of unknown environments with unmanned ground vehicles compared to the current state of the art. Its intended usage is in rescue scenarios in large indoor and underground environments with limited teleoperation ability. Local and global sampling are used to improve the exploration efficiency for large environments. Nodes are selected as the next exploration goal based on a gain-cost ratio derived from the assumed 3D map coverage at the particular node and the distance to it. The proposed approach features a continuously-built graph with a decoupled calculation of node gains using a computationally efficient ray tracing method. The Next-Best View is evaluated while the robot is pursuing a goal, which eliminates the need to wait for gain calculation after reaching the previous goal and significantly speeds up the exploration. Furthermore, a grid map is used to determine the traversability between the nodes in the graph while also providing a global plan for navigating towards selected goals. Simulations compare the proposed approach to state-of-the-art exploration algorithms and demonstrate its superior performance.

cs.RO↗

arXiv2021

SleepTransformer: Automatic Sleep Staging with Interpretability and Uncertainty Quantification

Background: Black-box skepticism is one of the main hindrances impeding deep-learning-based automatic sleep scoring from being used in clinical environments. Methods: Towards interpretability, this work proposes a sequence-to-sequence sleep-staging model, namely SleepTransformer. It is based on the transformer backbone and offers interpretability of the model's decisions at both the epoch and sequence level. We further propose a simple yet efficient method to quantify uncertainty in the model's decisions. The method, which is based on entropy, can serve as a metric for deferring low-confidence epochs to a human expert for further inspection. Results: Making sense of the transformer's self-attention scores for interpretability, at the epoch level, the attention scores are encoded as a heat map to highlight sleep-relevant features captured from the input EEG signal. At the sequence level, the attention scores are visualized as the influence of different neighboring epochs in an input sequence (i.e. the context) to recognition of a target epoch, mimicking the way manual scoring is done by human experts. Conclusion: Additionally, we demonstrate that SleepTransformer performs on par with existing methods on two databases of different sizes. Significance: Equipped with interpretability and the ability of uncertainty quantification, SleepTransformer holds promise for being integrated into clinical settings.

cs.LG↗

arXiv2021

Multi-view Audio and Music Classification

We propose in this work a multi-view learning approach for audio and music classification. Considering four typical low-level representations (i.e. different views) commonly used for audio and music recognition tasks, the proposed multi-view network consists of four subnetworks, each handling one input types. The learned embedding in the subnetworks are then concatenated to form the multi-view embedding for classification similar to a simple concatenation network. However, apart from the joint classification branch, the network also maintains four classification branches on the single-view embedding of the subnetworks. A novel method is then proposed to keep track of the learning behavior on the classification branches and adapt their weights to proportionally blend their gradients for network training. The weights are adapted in such a way that learning on a branch that is generalizing well will be encouraged whereas learning on a branch that is overfitting will be slowed down. Experiments on three different audio and music classification tasks show that the proposed multi-view network not only outperforms the single-view baselines but also is superior to the multi-view baselines based on concatenation and late fusion.

cs.SD↗

arXiv2020

Self-Attention Generative Adversarial Network for Speech Enhancement

Existing generative adversarial networks (GANs) for speech enhancement solely rely on the convolution operation, which may obscure temporal dependencies across the sequence input. To remedy this issue, we propose a self-attention layer adapted from non-local attention, coupled with the convolutional and deconvolutional layers of a speech enhancement GAN (SEGAN) using raw signal input. Further, we empirically study the effect of placing the self-attention layer at the (de)convolutional layers with varying layer indices as well as at all of them when memory allows. Our experiments show that introducing self-attention to SEGAN leads to consistent improvement across the objective evaluation metrics of enhancement performance. Furthermore, applying at different (de)convolutional layers does not significantly alter performance, suggesting that it can be conveniently applied at the highest-level (de)convolutional layer with the smallest memory overhead.

cs.SD↗

arXiv2020

A test for Heckscher-Ohlin using value-added exports

Empirical evidence for the Heckscher-Ohlin model has been inconclusive. We test whether the predictions of the Heckscher-Ohlin Theorem with respect to labor and capital find support in value-added trade. Defining labor-capital intensities and endowments as the ratio of hours worked to the nominal capital stock, we find evidence against Heckscher-Ohlin. However, taking the ratio of total factor compensations, and thus accounting for differences in technologies, we find strong support for it. That is, labor-abundant countries tend to export value-added in goods of labor-intensive industries. Moreover, differentiating between broad industries, we find support for nine out of twelve industries.

econ.GN↗

arXiv2020

Economic Complexity and Growth: Can value-added exports better explain the link?

In economic literature, economic complexity is typically approximated on the basis of an economy's gross export structure. However, in times of ever increasingly integrated global value chains, gross exports may convey an inaccurate image of a country's economic performance since they also incorporate foreign value-added and double-counted exports. Thus, I introduce a new empirical approach approximating economic complexity based on a country's value-added export structure. This approach leads to substantially different complexity rankings compared to established metrics. Moreover, the explanatory power of GDP per capita growth rates for a sample of 40 lower-middle- to high-income countries is considerably higher, even if controlling for typical growth regression covariates.

econ.GN↗

arXiv2020

On Multitask Loss Function for Audio Event Detection and Localization

Audio event localization and detection (SELD) have been commonly tackled using multitask models. Such a model usually consists of a multi-label event classification branch with sigmoid cross-entropy loss for event activity detection and a regression branch with mean squared error loss for direction-of-arrival estimation. In this work, we propose a multitask regression model, in which both (multi-label) event detection and localization are formulated as regression problems and use the mean squared error loss homogeneously for model training. We show that the common combination of heterogeneous loss functions causes the network to underfit the data whereas the homogeneous mean squared error loss leads to better convergence and performance. Experiments on the development and validation sets of the DCASE 2020 SELD task demonstrate that the proposed system also outperforms the DCASE 2020 SELD baseline across all the detection and localization metrics, reducing the overall SELD error (the combined metric) by approximately 10% absolute.

eess.AS↗

arXiv2020

XSleepNet: Multi-View Sequential Model for Automatic Sleep Staging

Automating sleep staging is vital to scale up sleep assessment and diagnosis to serve millions experiencing sleep deprivation and disorders and enable longitudinal sleep monitoring in home environments. Learning from raw polysomnography signals and their derived time-frequency image representations has been prevalent. However, learning from multi-view inputs (e.g., both the raw signals and the time-frequency images) for sleep staging is difficult and not well understood. This work proposes a sequence-to-sequence sleep staging model, XSleepNet, that is capable of learning a joint representation from both raw signals and time-frequency images. Since different views may generalize or overfit at different rates, the proposed network is trained such that the learning pace on each view is adapted based on their generalization/overfitting behavior. In simple terms, the learning on a particular view is speeded up when it is generalizing well and slowed down when it is overfitting. View-specific generalization/overfitting measures are computed on-the-fly during the training course and used to derive weights to blend the gradients from different views. As a result, the network is able to retain the representation power of different views in the joint features which represent the underlying distribution better than those learned by each individual view alone. Furthermore, the XSleepNet architecture is principally designed to gain robustness to the amount of training data and to increase the complementarity between the input views. Experimental results on five databases of different sizes show that XSleepNet consistently outperforms the single-view baselines and the multi-view baseline with a simple fusion strategy. Finally, XSleepNet also outperforms prior sleep staging methods and improves previous state-of-the-art results on the experimental databases.

eess.SP↗

arXiv2020

Personalized Automatic Sleep Staging with Single-Night Data: a Pilot Study with KL-Divergence Regularization

Brain waves vary between people. An obvious way to improve automatic sleep staging for longitudinal sleep monitoring is personalization of algorithms based on individual characteristics extracted from the first night of data. As a single night is a very small amount of data to train a sleep staging model, we propose a Kullback-Leibler (KL) divergence regularized transfer learning approach to address this problem. We employ the pretrained SeqSleepNet (i.e. the subject independent model) as a starting point and finetune it with the single-night personalization data to derive the personalized model. This is done by adding the KL divergence between the output of the subject independent model and the output of the personalized model to the loss function during finetuning. In effect, KL-divergence regularization prevents the personalized model from overfitting to the single-night data and straying too far away from the subject independent model. Experimental results on the Sleep-EDF Expanded database with 75 subjects show that sleep staging personalization with a single-night data is possible with help of the proposed KL-divergence regularization. On average, we achieve a personalized sleep staging accuracy of 79.6%, a Cohen's kappa of 0.706, a macro F1-score of 73.0%, a sensitivity of 71.8%, and a specificity of 94.2%. We find both that the approach is robust against overfitting and that it improves the accuracy by 4.5 percentage points compared to non-personalization and 2.2 percentage points compared to personalization without regularization.

cs.LG↗

arXiv2020

Improving GANs for Speech Enhancement

Generative adversarial networks (GAN) have recently been shown to be efficient for speech enhancement. However, most, if not all, existing speech enhancement GANs (SEGAN) make use of a single generator to perform one-stage enhancement mapping. In this work, we propose to use multiple generators that are chained to perform multi-stage enhancement mapping, which gradually refines the noisy input signals in a stage-wise fashion. Furthermore, we study two scenarios: (1) the generators share their parameters and (2) the generators' parameters are independent. The former constrains the generators to learn a common mapping that is iteratively applied at all enhancement stages and results in a small model footprint. On the contrary, the latter allows the generators to flexibly learn different enhancement mappings at different stages of the network at the cost of an increased model size. We demonstrate that the proposed multi-stage enhancement approach outperforms the one-stage SEGAN baseline, where the independent generators lead to more favorable results than the tied generators. The source code is available at http://github.com/pquochuy/idsegan.

cs.LG↗