ERC FUNDED PROJECTS

Rates of domestic violence and the relative risk of premature death for women are higher in sub-Saharan Africa than in any other region. Yet we know remarkably little about the economic forces, incentives and constraints that drive discrimination against women in this region, making it hard to identify policy levers to address the problem. This project will help fill this gap. I will investigate gender discrimination from two complementary perspectives. First, through the lens of economic history, I will investigate the forces driving trends in women’s relative well-being since slavery. To quantify the evolution of well-being of sub-Saharan women relative to men, I will use three types of historical data: anthropometric indicators (relative height), vital statistics (to compute numbers of missing women), and outcomes of formal and informal family law disputes. I will then investigate how major economic developments and changes in family laws differentially affected women’s welfare across ethnic groups with different norms on women’s roles and rights. Second, using intra-household economic models, I will provide new insights into domestic violence and gender bias in access to crucial resources in present-day Africa. I will develop a new household model that incorporates gender identity and endogenous outside options to explore the relationship between women’s empowerment and the use of violence. Using the notion of strategic delegation, I will propose a new rationale for the separation of budgets often observed in African households and generate predictions of how improvements in women’s outside options affect welfare. Finally, with first hand data, I will investigate intra-household differences in nutrition and work effort in times of food shortage from the points of view of efficiency and equity. I will use activity trackers as an innovative means of collecting high quality data on work effort and thus overcome data limitations restricting the existing literature

1 499 313 €

Start date: 2018-08-01, End date: 2023-07-31

Anxiety, a state of apprehension or fear, may provoke cognitive or behavioural disorders and eventually lead to serious medical illnesses. The high prevalence of anxiety disorders in our society sharply contrasts with the lack of clear factual knowledge about the corresponding brain mechanisms at the origin of this profound change in the appraisal of the environment. Little is known about how the psychopathological state of anxiety ultimately turns to a medical condition. The core of this proposal is to gain insight in the neural underpinnings of anxiety and disorders related to anxiety using modern human brain-imaging such as scalp EEG and fMRI. I propose to enlighten how anxiety transforms and shapes human cognition and what the neural correlates and time-course of this modulatory effect are. The primary innovation of this project is the systematic use scalp EEG and fMRI in human participants to better understand the neural mechanisms by which anxiety profoundly influences specific cognitive functions, in particular selective attention and decision-making. The goal of this proposal is to precisely determine the exact timing (using scalp EEG), location, size and extent (using fMRI) of anxiety-related modulations on selective attention and decision-making in the human brain. Here I propose to focus on these two specific processes, because they are likely to reveal selective effects of anxiety on human cognition and can thus serve as powerful models to better figure out how anxiety operates in the human brain. Another important aspect of this project is the fact I envision to help bridge the gap in Health Psychology between fundamental research and clinical practice by proposing alternative revalidation strategies for human adult subjects affected by anxiety-related disorders, which could directly exploit the neuro-scientific discoveries generated in this scientific project.

812 986 €

Start date: 2008-11-01, End date: 2013-10-31

The Bantu Expansion is not only the main linguistic, cultural and demographic process in Late Holocene Africa. It is also one of the most controversial issues in African History that still has political repercussions today. It has sparked debate across the disciplines and far beyond Africanist circles in an attempt to understand how the young Bantu language family (ca. 5000 years) could spread over large parts of Central, Eastern and Southern Africa. This massive dispersal is commonly seen as the result of a single migratory macro-event driven by agriculture, but many questions about the movement and subsistence of ancestral Bantu speakers are still open. They can only be answered through real interdisciplinary collaboration. This project will unite researchers with outstanding expertise in African archaeology, archaeobotany and historical linguistics to form a unique cross-disciplinary team that will shed new light on the first Bantu-speaking village communities south of the rainforest. Fieldwork is planned in parts of the Democratic Republic of Congo, the Republic of Congo and Angola that are terra incognita for archaeologists to determine the timing, location and archaeological signature of the earliest villagers and to establish how they interacted with autochthonous hunter-gatherers. Special attention will be paid to archaeobotanical and palaeoenvironmental data to get an idea of their subsistence, diet and habitat. Historical linguistics will be pushed beyond the boundaries of vocabulary-based phylogenetics and open new pathways in lexical reconstruction, especially regarding subsistence and land use of early Bantu speakers. Through interuniversity collaboration archaeozoological, palaeoenvironmental and genetic data and phylogenetic modelling will be brought into the cross-disciplinary approach to acquire a new holistic view on the interconnections between human migration, language spread, climate change and early farming in Late Holocene Central Africa.

1 997 500 €

Start date: 2018-01-01, End date: 2022-12-31

"Summary: the quest for a general functional tensor framework for blind source separation Our overall objective is the development of a general functional framework for solving tensor based blind source separation (BSS) problems in biomedical data fusion, using tensor decompositions (TDs) as basic core. We claim that TDs will allow the extraction of fairly complicated sources of biomedical activity from fairly complicated sets of uni- and multimodal data. The power of the new techniques will be demonstrated for three well-chosen representative biomedical applications for which extensive expertise and fully validated datasets are available in the PI’s team, namely: • Metabolite quantification and brain tumour tissue typing using Magnetic Resonance Spectroscopic Imaging, • Functional monitoring including seizure detection and polysomnography, • Cognitive brain functioning and seizure zone localization using simultaneous Electroencephalography-functional MR Imaging integration. Solving these challenging problems requires that algorithmic progress is made in several directions: • Algorithms need to be based on multilinear extensions of numerical linear algebra. • New grounds for separation, such as representability in a given function class, need to be explored. • Prior knowledge needs to be exploited via appropriate health relevant constraints. • Biomedical data fusion requires the combination of TDs, coupled via relevant constraints. • Algorithms for TD updating are important for continuous long-term patient monitoring. The algorithms are eventually integrated in an easy-to-use open source software platform that is general enough for use in other BSS applications. Having been involved in biomedical signal processing over a period of 20 years, the PI has a good overview of the field and the opportunities. By working directly at the forefront in close collaboration with the clinical scientists who actually use our software, we can have a huge impact."

2 500 000 €

Start date: 2014-04-01, End date: 2019-03-31

Natural language understanding (NLU) by the machine is of large scientific, economic and social value. Humans perform the NLU task in an efficient way by relying on their capability to imagine or anticipate situations. They engage commonsense and world knowledge that is often acquired through perceptual experiences to make explicit what is left implicit in language. Inspired by these characteristics CALCULUS will design, implement and evaluate innovative paradigms supporting NLU, where it will combine old but powerful ideas for language understanding from the early days of artificial intelligence with new approaches from machine learning. The project focuses on the effective learning of anticipatory, continuous, non-symbolic representations of event frames and narrative structures of events that are trained on language and visual data. The grammatical structure of language is grounded in the geometric structure of visual data while embodying aspects of commonsense and world knowledge. The reusable representations are evaluated in a selection of NLU tasks requiring efficient real-time retrieval of the representations and parsing of the targeted written texts. Finally, we will evaluate the inference potential of the anticipatory representations in situations not seen in the training data and when inferring spatial and temporal information in metric real world spaces that is not mentioned in the processed language. The machine learning methods focus on learning latent variable models relying on Bayesian probabilistic models and neural networks and focus on settings with limited training data that are manually annotated. The best models will be integrated in a demonstrator that translates the language of stories to events happening in a 3-D virtual world. The PI has interdisciplinary expertise in natural language processing, joint processing of language and visual data, information retrieval and machine learning needed for the successful realization of the project.

2 227 500 €

Start date: 2018-09-01, End date: 2023-08-31

China s economy has expanded at breakneck speed to become the 3rd largest trading country in the world and the largest recipient of foreign direct investment (FDI). Entry into the WTO in 2001 was a landmark event in this ongoing process and I propose to study several channels through which it spurred China s industrial development. Crucially, I will take an integrated view of the different ways in which Chinese and Western firms interact: through trade flows, as suppliers or competitors, FDI, or knowledge transfers. First, I investigate the existence and magnitude of a causal link from the trade reforms to productivity growth. Second, I look for evidence of capability upgrading, such as increased production efficiency, an ability to produce higher quality products, or introduce new products by innovating. Third, I study the mechanisms for the impact of trade and FDI on local firms, in particular assessing the relative importance of increased market competition and the transfer of know-how from foreign firms. For this analysis, I draw heavily on a unique data set. Information on the universe of Chinese manufacturing firms is being linked to the universe of Chinese trade transactions. These are unique research tools on their own, but as a linked data set, the only comparable one in the world is for the U.S. economy. The Chinese data has the advantage to contain detailed information on FDI, distinguishes between ordinary and processing trade, and contains information on innovation, such as R&D and sales of new goods. Answering the above questions is important for other developing countries wanting to learn from China s experience and for Western firms assessing how quickly Chinese firms will become viable suppliers of sophisticated inputs or direct competitors. By estimating models that are explicitly derived from new theories, I advance the literature at the interaction of international and development economics, industrial organization, economic geography.

944 940 €

Start date: 2010-02-01, End date: 2016-01-31

One of the primary and most appealing goals of computer vision is to automatically understand the content of images on a cognitive level. Ultimately we want to have computers interpret images as we humans do, recognizing all the objects, scenes, and people as well as their relations as they appear in natural images or video. With this project, I want to advance the state of the art in this field in two directions, which I believe to be crucial to build the next generation of image understanding tools. First, novel more robust yet descriptive image representations will be designed, that incorporate the intrinsic structure of images. These should already go a long way towards removing irrelevant sources of variability while capturing the essence of the image content. I believe the importance of further research into image representations is currently underestimated within the research community, yet I claim this is a crucial step with lots of opportunities good learning cannot easily make up for bad features. Second, weakly supervised methods to learn from multimodal input (especially the combination of images and text) will be investigated, making it possible to leverage the large amount of weak annotations available via the internet. This is essential if we want to scale the methods to a larger number of object categories (several hundreds instead of a few tens). As more data can be used for training, such weakly supervised methods might in the end even come on par with or outperform supervised schemes. Here we will call upon the latest results in semi-supervised learning, datamining, and computational linguistics.

1 538 380 €

Start date: 2010-02-01, End date: 2015-01-31

Low-rank matrix approximation (LRA) techniques such as principal component analysis (PCA) are powerful tools for the representation and analysis of high dimensional data, and are used in a wide variety of areas such as machine learning, signal and image processing, data mining, and optimization. Without any constraints and using the least squares error, LRA can be solved via the singular value decomposition. However, in practice, this model is often not suitable mainly because (i) the data might be contaminated with outliers, missing data and non-Gaussian noise, and (ii) the low-rank factors of the decomposition might have to satisfy some specific constraints. Hence, in recent years, many variants of LRA have been introduced, using different constraints on the factors and using different objective functions to assess the quality of the approximation; e.g., sparse PCA, PCA with missing data, independent component analysis and nonnegative matrix factorization. Although these new constrained LRA models have become very popular and standard in some fields, there is still a significant gap between theory and practice. In this project, our goal is to reduce this gap by attacking the problem in an integrated way making connections between LRA variants, and by using four very different but complementary perspectives: (1) computational complexity issues, (2) provably correct algorithms, (3) heuristics for difficult instances, and (4) application-oriented aspects. This unified and multi-disciplinary approach will enable us to understand these problems better, to develop and analyze new and existing algorithms and to then use them for applications. Our ultimate goal is to provide practitioners with new tools and to allow them to decide which method to use in which situation and to know what to expect from it.

1 291 750 €

Start date: 2016-09-01, End date: 2021-08-31

Side-channel attacks are an important threat against cryptographic implementations in which an adversary takes advantage of physical leakages, such as the power consumption of a smart card, in order to recover secret information. By circumventing the models in which standard security proofs are obtained, they can lead to powerful attacks against a large class of devices. As a consequence, formalizing implementation security and efficiently preventing side-channel attacks is one of the most challenging open problems in modern cryptography. Physical attacks imply new optimization criteria, with potential impact on the way we conceive algorithms and the way we design circuits. By putting together mathematical and electrical engineering problems, just as they are raised in reality, the CRASH project is expected to develop concrete basements for the next generation of cryptographic algorithms and their implementation. For this purpose, three main directions will be considered. First, we will investigate sound evaluation tools for side-channel attacks and validate them on different prototype chips. Second, we will consider the impact of physical attacks on the mathematical aspects of cryptography, both destructively (i.e. by developing new attacks and advanced cryptanalysis tools) and constructively (i.e. by investigating new cipher designs and security proof techniques). Third, we will evaluate the possibility to integrate physical security analysis into the design tools of integrated circuits (e.g. in order to obtain “physical security aware” compilers). Summarizing, this project aims to break the barrier between the abstractions of mathematical cryptography and the concrete peculiarities of physical security in present microelectronic devices. By considering the system and algorithmic issues in a unified way, it is expected to get rid of the incompatibilities between the separate formalisms that are usually considered in order to explain these concurrent realities.

1 498 874 €

Start date: 2011-10-01, End date: 2016-09-30