Project acronym BEHAVFRICTIONS
Project Behavioral Implications of Information-Processing Frictions
Researcher (PI) Jakub STEINER
Host Institution (HI) NARODOHOSPODARSKY USTAV AKADEMIE VED CESKE REPUBLIKY VEREJNA VYZKUMNA INSTITUCE
Country Czechia
Call Details Consolidator Grant (CoG), SH1, ERC-2017-COG
Summary BEHAVFRICTIONS will use novel models focussing on information-processing frictions to explain choice patterns described in behavioral economics and psychology. The proposed research will provide microfoundations that are essential for (i) identification of stable preferences, (ii) counterfactual predictions, and (iii) normative conclusions.
(i) Agents who face information-processing costs must trade the precision of choice against information costs. Their behavior thus reflects both their stable preferences and the context-dependent procedures that manage their errors stemming from imperfect information processing. In the absence of micro-founded models, the two drivers of the behavior are difficult to disentangle for outside observers. In some pillars of the proposal, the agents follow choice rules that closely resemble logit rules used in structural estimation. This will allow me to reinterpret the structural estimation fits to choice data and to make a distinction between the stable preferences and frictions.
(ii) Such a distinction is important in counterfactual policy analysis because the second-best decision procedures that manage the errors in choice are affected by the analysed policy. Incorporation of the information-processing frictions into existing empirical methods will improve our ability to predict effects of the policies.
(iii) My preliminary results suggest that when an agent is prone to committing errors, biases--such as overconfidence, confirmatory bias, or perception biases known from prospect theory--arise under second-best strategies. By providing the link between the agent's environment and the second-best distribution of the perception errors, my models will delineate environments in which these biases shield the agents from the most costly mistakes from environments in which the biases turn into maladaptations. The distinction will inform the normative debate on debiasing.
Summary
BEHAVFRICTIONS will use novel models focussing on information-processing frictions to explain choice patterns described in behavioral economics and psychology. The proposed research will provide microfoundations that are essential for (i) identification of stable preferences, (ii) counterfactual predictions, and (iii) normative conclusions.
(i) Agents who face information-processing costs must trade the precision of choice against information costs. Their behavior thus reflects both their stable preferences and the context-dependent procedures that manage their errors stemming from imperfect information processing. In the absence of micro-founded models, the two drivers of the behavior are difficult to disentangle for outside observers. In some pillars of the proposal, the agents follow choice rules that closely resemble logit rules used in structural estimation. This will allow me to reinterpret the structural estimation fits to choice data and to make a distinction between the stable preferences and frictions.
(ii) Such a distinction is important in counterfactual policy analysis because the second-best decision procedures that manage the errors in choice are affected by the analysed policy. Incorporation of the information-processing frictions into existing empirical methods will improve our ability to predict effects of the policies.
(iii) My preliminary results suggest that when an agent is prone to committing errors, biases--such as overconfidence, confirmatory bias, or perception biases known from prospect theory--arise under second-best strategies. By providing the link between the agent's environment and the second-best distribution of the perception errors, my models will delineate environments in which these biases shield the agents from the most costly mistakes from environments in which the biases turn into maladaptations. The distinction will inform the normative debate on debiasing.
Max ERC Funding
1 321 488 €
Duration
Start date: 2018-06-01, End date: 2023-05-31
Project acronym ChinaCreative
Project From Made in China to Created in China - A Comparative Study of Creative Practice and Production in Contemporary China
Researcher (PI) Bastiaan Jeroen De Kloet
Host Institution (HI) UNIVERSITEIT VAN AMSTERDAM
Country Netherlands
Call Details Consolidator Grant (CoG), SH5, ERC-2013-CoG
Summary With its emergence as a global power, China aspires to move from a “made in China” towards a “created in China” country. Creativity and culture have become a crucial source for innovation and financial growth, but are also mobilised to promote a new and open China to both the citizenry as well as the outside world. They are part of what is termed China’s “soft power.”
What does creativity mean in the context of China, and what does it do? When both the state and profoundly globalised creative industries are so deeply implicated in the promotion of creativity, what are the possibilities of criticality, if any? Whereas creativity has been extensively researched in the fields of psychology, law and neurosciences, scholarship in the humanities has by and large side-tracked the thorny issue of creativity. Yet, the worldwide resurgence of the term under the banner of creative industries makes it all the more urgent to develop a theory of creativity. This project understands creativity as a textual, a social as well as a heritage practice. It aims to analyse claims of creativity in different cultural practices, and to analyse how emerging creativities in China are part of tactics of governmentality and disable or enable possibilities of criticality.
Using a comparative, multi-disciplinary, multi-method and multi-sited research design, five subprojects analyse (1) contemporary art, (2) calligraphy, (3) independent documentary cinema, (4) television from Hunan Satellite TV and (5) “fake” (shanzhai) art. By including both popular and high arts, by including both more Westernized as well as more specifically Chinese art forms, by including both the “real” as well as the “fake,” by studying different localities, and by mobilising methods from both the social sciences and the humanities, this project is pushing the notion of comparative research to a new level.
Summary
With its emergence as a global power, China aspires to move from a “made in China” towards a “created in China” country. Creativity and culture have become a crucial source for innovation and financial growth, but are also mobilised to promote a new and open China to both the citizenry as well as the outside world. They are part of what is termed China’s “soft power.”
What does creativity mean in the context of China, and what does it do? When both the state and profoundly globalised creative industries are so deeply implicated in the promotion of creativity, what are the possibilities of criticality, if any? Whereas creativity has been extensively researched in the fields of psychology, law and neurosciences, scholarship in the humanities has by and large side-tracked the thorny issue of creativity. Yet, the worldwide resurgence of the term under the banner of creative industries makes it all the more urgent to develop a theory of creativity. This project understands creativity as a textual, a social as well as a heritage practice. It aims to analyse claims of creativity in different cultural practices, and to analyse how emerging creativities in China are part of tactics of governmentality and disable or enable possibilities of criticality.
Using a comparative, multi-disciplinary, multi-method and multi-sited research design, five subprojects analyse (1) contemporary art, (2) calligraphy, (3) independent documentary cinema, (4) television from Hunan Satellite TV and (5) “fake” (shanzhai) art. By including both popular and high arts, by including both more Westernized as well as more specifically Chinese art forms, by including both the “real” as well as the “fake,” by studying different localities, and by mobilising methods from both the social sciences and the humanities, this project is pushing the notion of comparative research to a new level.
Max ERC Funding
1 947 448 €
Duration
Start date: 2014-09-01, End date: 2019-08-31
Project acronym CoAct
Project Communication in Action: Towards a model of Contextualized Action and Language Processing
Researcher (PI) Judith HOLLER
Host Institution (HI) STICHTING KATHOLIEKE UNIVERSITEIT
Country Netherlands
Call Details Consolidator Grant (CoG), SH4, ERC-2017-COG
Summary Language is fundamental to human sociality. While the last century has given us many fundamental insights into how we use and understand it, core issues that we face when doing so within its natural environment—face-to-face conversation—remain untackled. When we speak we also send signals with our head, eyes, face, hands, torso, etc. How do we orchestrate and integrate all this information into meaningful messages? CoAct will lead to a new model with in situ language processing at its core, the Contextualized Action and Language (CoALa) processing model. The defining characteristic of in situ language is its multimodal nature. Moreover, the essence of language use is social action; that is, we use language to do things—we question, offer, decline etc. These social actions are embedded in conversational structure where one speaking turn follows another at a remarkable speed, with millisecond gaps between them. Conversation thus confronts us with a significant psycholinguistic challenge. While one could expect that the many co-speech bodily signals exacerbate this challenge, CoAct proposes that they actually play a key role in dealing with it. It tests this in three subprojects that combine methods from a variety of disciplines but focus on the social actions performed by questions and responses as a uniting theme: (1) ProdAct uses conversational corpora to investigate the multimodal architecture of social actions with the assumption that they differ in their ‘visual signatures’, (2) CompAct tests whether these bodily signatures contribute to social action comprehension, and if they do so early and rapidly, (3) IntAct investigates whether bodily signals play a facilitating role also when faced with the complex task of comprehending while planning a next social action. Thus, CoAct aims to advance current psycholinguistic theory by developing a new model of language processing based on an integrative framework uniting aspects from psychology , philosophy and sociology.
Summary
Language is fundamental to human sociality. While the last century has given us many fundamental insights into how we use and understand it, core issues that we face when doing so within its natural environment—face-to-face conversation—remain untackled. When we speak we also send signals with our head, eyes, face, hands, torso, etc. How do we orchestrate and integrate all this information into meaningful messages? CoAct will lead to a new model with in situ language processing at its core, the Contextualized Action and Language (CoALa) processing model. The defining characteristic of in situ language is its multimodal nature. Moreover, the essence of language use is social action; that is, we use language to do things—we question, offer, decline etc. These social actions are embedded in conversational structure where one speaking turn follows another at a remarkable speed, with millisecond gaps between them. Conversation thus confronts us with a significant psycholinguistic challenge. While one could expect that the many co-speech bodily signals exacerbate this challenge, CoAct proposes that they actually play a key role in dealing with it. It tests this in three subprojects that combine methods from a variety of disciplines but focus on the social actions performed by questions and responses as a uniting theme: (1) ProdAct uses conversational corpora to investigate the multimodal architecture of social actions with the assumption that they differ in their ‘visual signatures’, (2) CompAct tests whether these bodily signatures contribute to social action comprehension, and if they do so early and rapidly, (3) IntAct investigates whether bodily signals play a facilitating role also when faced with the complex task of comprehending while planning a next social action. Thus, CoAct aims to advance current psycholinguistic theory by developing a new model of language processing based on an integrative framework uniting aspects from psychology , philosophy and sociology.
Max ERC Funding
2 000 000 €
Duration
Start date: 2018-09-01, End date: 2023-08-31
Project acronym CONT-END
Project Attempts to Control the End of Life in People with Dementia: Two-level Approach to Examine Controversies
Researcher (PI) Jenny VAN DER STEEN
Host Institution (HI) ACADEMISCH ZIEKENHUIS LEIDEN
Country Netherlands
Call Details Consolidator Grant (CoG), SH3, ERC-2017-COG
Summary In dementia at the end of life, cognitive and physical decline imply that control is typically lost. CONT-END will examine control in the context of three emerging interventions which contain a controversial element of striving for control in the process of dying with dementia: advance care planning of the end of life, use of new technology to monitor symptoms when unable to self-report, and euthanasia. To perform outstanding research, the proposed research examines control at the level of clinical practice, but also at the level of end-of-life research practice. The latter provides ample opportunities for researchers to control the research process. That is, research designs are often flexible and we will study if and how this impacts research in an emotionally charged area. I will take an empirical mixed-methods approach to study the two practices in parallel. The work is organised in three related Work Packages around three research questions. (1) In a 6-country study, I will examine if and when people with dementia, family caregivers and physicians (900 respondents) find the interventions, shown on video, acceptable. (2) A cluster-randomised 3-armed controlled trial in 279 patients and their family caregivers assesses effects of two types of advance care planning differing in level of control (detailed advance treatment orders versus goal setting and coping based) on outcomes ranging from favourable to less favourable, and whether effects differ in subgroups. Cases in which the technology is preferred or applied are observed. (3) Ethnographic fieldwork in two different end-of-life research practices and a Delphi study to synthesize CONT-END’s findings assess how researchers shape findings. This greatly improves the quality of CONT-END and provides the input to develop new methodology for improving research quality and integrity.
Summary
In dementia at the end of life, cognitive and physical decline imply that control is typically lost. CONT-END will examine control in the context of three emerging interventions which contain a controversial element of striving for control in the process of dying with dementia: advance care planning of the end of life, use of new technology to monitor symptoms when unable to self-report, and euthanasia. To perform outstanding research, the proposed research examines control at the level of clinical practice, but also at the level of end-of-life research practice. The latter provides ample opportunities for researchers to control the research process. That is, research designs are often flexible and we will study if and how this impacts research in an emotionally charged area. I will take an empirical mixed-methods approach to study the two practices in parallel. The work is organised in three related Work Packages around three research questions. (1) In a 6-country study, I will examine if and when people with dementia, family caregivers and physicians (900 respondents) find the interventions, shown on video, acceptable. (2) A cluster-randomised 3-armed controlled trial in 279 patients and their family caregivers assesses effects of two types of advance care planning differing in level of control (detailed advance treatment orders versus goal setting and coping based) on outcomes ranging from favourable to less favourable, and whether effects differ in subgroups. Cases in which the technology is preferred or applied are observed. (3) Ethnographic fieldwork in two different end-of-life research practices and a Delphi study to synthesize CONT-END’s findings assess how researchers shape findings. This greatly improves the quality of CONT-END and provides the input to develop new methodology for improving research quality and integrity.
Max ERC Funding
1 988 972 €
Duration
Start date: 2018-12-01, End date: 2023-11-30
Project acronym CORNEA
Project Controlling evolutionary dynamics of networked autonomous agents
Researcher (PI) Ming CAO
Host Institution (HI) RIJKSUNIVERSITEIT GRONINGEN
Country Netherlands
Call Details Consolidator Grant (CoG), PE7, ERC-2017-COG
Summary Large-scale technological, biological, economic, and social complex systems act as complex networks of interacting autonomous agents. Large numbers of interacting agents making self-interested decisions can result in highly complex, sometimes surprising, and often suboptimal, collective behaviors. Empowered by recent breakthroughs in data-driven cognitive learning technologies, networked agents collectively give rise to evolutionary dynamics that cannot be easily modeled, analysed and/or controlled using current systems and control theory. Consequently, there is an urgent need to develop new theoretical foundations to tackle the emerging challenging control problems associated with evolutionary dynamics for networked autonomous agents.
The aim of this project is to develop a rigorous theory for the control of evolutionary dynamics so that interacting autonomous agents can be guided to solve group tasks through the pursuit of individual goals in an evolutionary dynamical process. The theory will then be tested, validated and improved against experimental results using robotic fish.
To achieve the aim, I will: (1) develop a general formulation for stochastic evolutionary dynamics with control inputs, enabling the study on controllability and stabilizability for evolutionary processes; (2) introduce stochastic control Lyapunov functions to design control laws; (3) construct new classes of conditional strategies that may propagate controlled actions effectively from focal agents in multiple time scales; and (4) validate experimentally on tasks with unknown difficulties that require a group of robotic fish to evolve and adapt.
The project will result in a major advance from the conventional usage of evolutionary game theory with the systematic design to actively control evolutionary outcomes. The combination of theory with experimentation and the multi-disciplinary nature of the approach will lead to new applications of autonomous robotic systems.
Summary
Large-scale technological, biological, economic, and social complex systems act as complex networks of interacting autonomous agents. Large numbers of interacting agents making self-interested decisions can result in highly complex, sometimes surprising, and often suboptimal, collective behaviors. Empowered by recent breakthroughs in data-driven cognitive learning technologies, networked agents collectively give rise to evolutionary dynamics that cannot be easily modeled, analysed and/or controlled using current systems and control theory. Consequently, there is an urgent need to develop new theoretical foundations to tackle the emerging challenging control problems associated with evolutionary dynamics for networked autonomous agents.
The aim of this project is to develop a rigorous theory for the control of evolutionary dynamics so that interacting autonomous agents can be guided to solve group tasks through the pursuit of individual goals in an evolutionary dynamical process. The theory will then be tested, validated and improved against experimental results using robotic fish.
To achieve the aim, I will: (1) develop a general formulation for stochastic evolutionary dynamics with control inputs, enabling the study on controllability and stabilizability for evolutionary processes; (2) introduce stochastic control Lyapunov functions to design control laws; (3) construct new classes of conditional strategies that may propagate controlled actions effectively from focal agents in multiple time scales; and (4) validate experimentally on tasks with unknown difficulties that require a group of robotic fish to evolve and adapt.
The project will result in a major advance from the conventional usage of evolutionary game theory with the systematic design to actively control evolutionary outcomes. The combination of theory with experimentation and the multi-disciplinary nature of the approach will lead to new applications of autonomous robotic systems.
Max ERC Funding
1 998 933 €
Duration
Start date: 2018-05-01, End date: 2023-04-30
Project acronym EXPECT HEAL-TH
Project Empowering expectations for health and disease: training the immune and endocrine system
Researcher (PI) Andrea Evers
Host Institution (HI) UNIVERSITEIT LEIDEN
Country Netherlands
Call Details Consolidator Grant (CoG), SH4, ERC-2013-CoG
Summary Expectations about health and disease induce immune and endocrine responses and directly affect health and treatment outcomes. However, there is an urge to understand the mechanical underpinnings how expectations affect immune and endocrine responses and how this knowledge can be used for therapeutic interventions.
My research group studies the main expectancy learning mechanisms for itch and pain as a generic expectancy model across symptoms and conditions. We recently showed that dual expectancy learning processes (i.e. conditioning and suggestions) are most powerful for itch symptoms, corresponding with findings for other symptoms and conditions. Based on these studies, I propose a groundbreaking dual expectancy learning approach, testing whether combined expectancy learning processes (i.e. both conditioning and suggestions, offered with personalized cues and exposure to relevant stressors) affect most profoundly the immune and endocrine system, in turn affecting health and disease outcomes.
The major aim is to unravel the central mechanisms of how peoples’ expectations affect immune and endocrine responses and related health outcomes, through the use of pioneering multidisciplinary methods in healthy and clinical populations. First, we systematically train immune and endocrine responses and relate them to psychological, neurobiological and genetic mechanisms. Second, we test these manipulations for physical health challenges (e.g. inflammatory or allergic histamine reactions) in healthy subjects and patients. Third, we study the long-term effects in chronic inflammatory itch and pain conditions (e.g. replacing anti-inflammatory pharmacotherapies, reducing side effects).
This interdisciplinary, cross-boundary project progresses key theoretical knowledge of the central expectation mechanisms for immune and endocrine responses. Findings open new horizons for health prevention and therapeutic interventions for various inflammatory conditions and physical symptoms.
Summary
Expectations about health and disease induce immune and endocrine responses and directly affect health and treatment outcomes. However, there is an urge to understand the mechanical underpinnings how expectations affect immune and endocrine responses and how this knowledge can be used for therapeutic interventions.
My research group studies the main expectancy learning mechanisms for itch and pain as a generic expectancy model across symptoms and conditions. We recently showed that dual expectancy learning processes (i.e. conditioning and suggestions) are most powerful for itch symptoms, corresponding with findings for other symptoms and conditions. Based on these studies, I propose a groundbreaking dual expectancy learning approach, testing whether combined expectancy learning processes (i.e. both conditioning and suggestions, offered with personalized cues and exposure to relevant stressors) affect most profoundly the immune and endocrine system, in turn affecting health and disease outcomes.
The major aim is to unravel the central mechanisms of how peoples’ expectations affect immune and endocrine responses and related health outcomes, through the use of pioneering multidisciplinary methods in healthy and clinical populations. First, we systematically train immune and endocrine responses and relate them to psychological, neurobiological and genetic mechanisms. Second, we test these manipulations for physical health challenges (e.g. inflammatory or allergic histamine reactions) in healthy subjects and patients. Third, we study the long-term effects in chronic inflammatory itch and pain conditions (e.g. replacing anti-inflammatory pharmacotherapies, reducing side effects).
This interdisciplinary, cross-boundary project progresses key theoretical knowledge of the central expectation mechanisms for immune and endocrine responses. Findings open new horizons for health prevention and therapeutic interventions for various inflammatory conditions and physical symptoms.
Max ERC Funding
1 981 009 €
Duration
Start date: 2014-09-01, End date: 2019-08-31
Project acronym FORCE-OF-GOSSIP
Project The unknown force: How gossip shapes the functioning and performance of organizational groups.
Researcher (PI) Bianca BEERSMA
Host Institution (HI) STICHTING VU
Country Netherlands
Call Details Consolidator Grant (CoG), SH3, ERC-2017-COG
Summary "Aim: To radically change the field of gossip research, this project builds theory that connects divergent perspectives on the effects of gossip on work group outcomes.
Background: Gossip, informal evaluative talk about absent third parties, is omnipresent in organizational groups. However, for scientists and practitioners alike, it is still an ""unknown force"" in organizations, because research findings about its effects on group functioning and performance are inconsistent. Whereas some studies point to the disruptive aspects of gossip, others claim it enhances group cooperation.
Innovation: I present the Gossip Origins, Subsequent Social Information Processing, and Performance (GOSSIPP)-framework, a new paradigm that views gossip in groups as a multi-level phenomenon: It, for the first time, systematically connects individual gossipers' intentions to group-level outcomes via social information processing by gossip recipients.
Propositions: Group members may gossip to benefit themselves or their group (proself vs. prosocial intentions, or motives). Receivers' reactions to gossip are affected by how they perceive senders' motives. These reactions in turn affect group-level processes. How these processes affect group performance depends on the type of group task.
Methodology: Four subprojects test the framework, applying a cross-disciplinary multi-method approach. Subprojects 1 and 2 employ laboratory experiments to assess causal effects of gossipers' motives and recipients' reactions on group processes and performance. Subproject 3 examines effects of different group compositions and feedback loops between performance and gossip, employing self-organizing computer modelling. Finally, to assess external validity, Subproject 4 examines work teams. Results will lead to long-sought understanding of when and why gossip is a functional or dysfunctional force and will enable evidence-based advice to organizations about the meaning and functionality of gossip for groups."
Summary
"Aim: To radically change the field of gossip research, this project builds theory that connects divergent perspectives on the effects of gossip on work group outcomes.
Background: Gossip, informal evaluative talk about absent third parties, is omnipresent in organizational groups. However, for scientists and practitioners alike, it is still an ""unknown force"" in organizations, because research findings about its effects on group functioning and performance are inconsistent. Whereas some studies point to the disruptive aspects of gossip, others claim it enhances group cooperation.
Innovation: I present the Gossip Origins, Subsequent Social Information Processing, and Performance (GOSSIPP)-framework, a new paradigm that views gossip in groups as a multi-level phenomenon: It, for the first time, systematically connects individual gossipers' intentions to group-level outcomes via social information processing by gossip recipients.
Propositions: Group members may gossip to benefit themselves or their group (proself vs. prosocial intentions, or motives). Receivers' reactions to gossip are affected by how they perceive senders' motives. These reactions in turn affect group-level processes. How these processes affect group performance depends on the type of group task.
Methodology: Four subprojects test the framework, applying a cross-disciplinary multi-method approach. Subprojects 1 and 2 employ laboratory experiments to assess causal effects of gossipers' motives and recipients' reactions on group processes and performance. Subproject 3 examines effects of different group compositions and feedback loops between performance and gossip, employing self-organizing computer modelling. Finally, to assess external validity, Subproject 4 examines work teams. Results will lead to long-sought understanding of when and why gossip is a functional or dysfunctional force and will enable evidence-based advice to organizations about the meaning and functionality of gossip for groups."
Max ERC Funding
1 999 693 €
Duration
Start date: 2018-09-01, End date: 2023-08-31
Project acronym FORSIED
Project Formalizing Subjective Interestingness in Exploratory Data Mining
Researcher (PI) Tijl De Bie
Host Institution (HI) UNIVERSITEIT GENT
Country Belgium
Call Details Consolidator Grant (CoG), PE6, ERC-2013-CoG
Summary "The rate at which research labs, enterprises and governments accumulate data is high and fast increasing. Often, these data are collected for no specific purpose, or they turn out to be useful for unanticipated purposes: Companies constantly look for new ways to monetize their customer databases; Governments mine various databases to detect tax fraud; Security agencies mine and cross-associate numerous heterogeneous information streams from publicly accessible and classified databases to understand and detect security threats. The objective in such Exploratory Data Mining (EDM) tasks is typically ill-defined, i.e. it is unclear how to formalize how interesting a pattern extracted from the data is. As a result, EDM is often a slow process of trial and error.
During this fellowship we aim to develop the mathematical principles of what makes a pattern interesting in a very subjective sense. Crucial in this endeavour will be research into automatic mechanisms to model and duly consider the prior beliefs and expectations of the user for whom the EDM patterns are intended, thus relieving the users of the complex task to attempt to formalize themselves what makes a pattern interesting to them.
This project will represent a radical change in how EDM research is done. Currently, researchers typically imagine a specific purpose for the patterns, try to formalize interestingness of such patterns given that purpose, and design an algorithm to mine them. However, given the variety of users, this strategy has led to a multitude of algorithms. As a result, users need to be data mining experts to understand which algorithm applies to their situation. To resolve this, we will develop a theoretically solid framework for the design of EDM systems that model the user's beliefs and expectations as much as the data itself, so as to maximize the amount of useful information transmitted to the user. This will ultimately bring the power of EDM within reach of the non-expert."
Summary
"The rate at which research labs, enterprises and governments accumulate data is high and fast increasing. Often, these data are collected for no specific purpose, or they turn out to be useful for unanticipated purposes: Companies constantly look for new ways to monetize their customer databases; Governments mine various databases to detect tax fraud; Security agencies mine and cross-associate numerous heterogeneous information streams from publicly accessible and classified databases to understand and detect security threats. The objective in such Exploratory Data Mining (EDM) tasks is typically ill-defined, i.e. it is unclear how to formalize how interesting a pattern extracted from the data is. As a result, EDM is often a slow process of trial and error.
During this fellowship we aim to develop the mathematical principles of what makes a pattern interesting in a very subjective sense. Crucial in this endeavour will be research into automatic mechanisms to model and duly consider the prior beliefs and expectations of the user for whom the EDM patterns are intended, thus relieving the users of the complex task to attempt to formalize themselves what makes a pattern interesting to them.
This project will represent a radical change in how EDM research is done. Currently, researchers typically imagine a specific purpose for the patterns, try to formalize interestingness of such patterns given that purpose, and design an algorithm to mine them. However, given the variety of users, this strategy has led to a multitude of algorithms. As a result, users need to be data mining experts to understand which algorithm applies to their situation. To resolve this, we will develop a theoretically solid framework for the design of EDM systems that model the user's beliefs and expectations as much as the data itself, so as to maximize the amount of useful information transmitted to the user. This will ultimately bring the power of EDM within reach of the non-expert."
Max ERC Funding
1 549 315 €
Duration
Start date: 2014-05-01, End date: 2019-04-30
Project acronym FoTran
Project Found in Translation – Natural Language Understanding with Cross-Lingual Grounding
Researcher (PI) Joerg TIEDEMANN
Host Institution (HI) HELSINGIN YLIOPISTO
Country Finland
Call Details Consolidator Grant (CoG), PE6, ERC-2017-COG
Summary "Natural language understanding is the ""holy grail"" of computational linguistics and a long-term goal in research on artificial intelligence. Understanding human communication is difficult due to the various ambiguities in natural languages and the wide range of contextual dependencies required to resolve them. Discovering the semantics behind language input is necessary for proper interpretation in interactive tools, which requires an abstraction from language-specific forms to language-independent meaning representations. With this project, I propose a line of research that will focus on the development of novel data-driven models that can learn such meaning representations from indirect supervision provided by human translations covering a substantial proportion of the linguistic diversity in the world. A guiding principle is cross-lingual grounding, the effect of resolving ambiguities through translation. The beauty of that idea is the use of naturally occurring data instead of artificially created resources and costly manual annotations. The framework is based on deep learning and neural machine translation and my hypothesis is that training on increasing amounts of linguistically diverse data improves the abstractions found by the model. Eventually, this will lead to universal sentence-level meaning representations and we will test our ideas with multilingual machine translation and tasks that require semantic reasoning and inference."
Summary
"Natural language understanding is the ""holy grail"" of computational linguistics and a long-term goal in research on artificial intelligence. Understanding human communication is difficult due to the various ambiguities in natural languages and the wide range of contextual dependencies required to resolve them. Discovering the semantics behind language input is necessary for proper interpretation in interactive tools, which requires an abstraction from language-specific forms to language-independent meaning representations. With this project, I propose a line of research that will focus on the development of novel data-driven models that can learn such meaning representations from indirect supervision provided by human translations covering a substantial proportion of the linguistic diversity in the world. A guiding principle is cross-lingual grounding, the effect of resolving ambiguities through translation. The beauty of that idea is the use of naturally occurring data instead of artificially created resources and costly manual annotations. The framework is based on deep learning and neural machine translation and my hypothesis is that training on increasing amounts of linguistically diverse data improves the abstractions found by the model. Eventually, this will lead to universal sentence-level meaning representations and we will test our ideas with multilingual machine translation and tasks that require semantic reasoning and inference."
Max ERC Funding
1 817 622 €
Duration
Start date: 2018-09-01, End date: 2023-08-31
Project acronym HELENA
Project Heavy-Element Nanowires
Researcher (PI) Erik Petrus Antonius Maria Bakkers
Host Institution (HI) TECHNISCHE UNIVERSITEIT DELFT
Country Netherlands
Call Details Consolidator Grant (CoG), PE5, ERC-2013-CoG
Summary "Nanowires are a powerful and versatile platform for a broad range of applications. Among all semiconductors, the heavy-elements materials exhibit the highest electron mobilities, strongest spin-orbit coupling and best thermoelectric properties. Nonetheless, heavy-element nanowires have been unexplored. With this proposal we unite the unique advantages of design freedom of nanowires with the special properties of heavy-element semiconductors. We specifically reveal the potential of heavy-element nanowires in the areas of thermoelectrics, and topological insulators. Using our strong track record in this area, we will pioneer the synthesis of this new class of materials and study their intrinsic materials properties. Starting point are nanowires of InSb and PbTe grown using the vapor-liquid-solid mechanism. Our aims are 1) to obtain highest-possible electron mobilities for these bottom-up fabricated materials by investigating new materials combinations of different semiconductor classes to effectively passivate the nanowire surface and we will eliminate impurities; 2) to investigate and optimize thermoelectric properties by developing advanced superlattice and core/shell nanowire structures where electronic and phononic transport is decoupled; and 3) to fabricate high-quality planar nanowire networks, which enable four-point electronic transport measurements and allow precisely determining carrier concentration and mobility. Besides the fundamentally interesting materials science, the heavy-element nanowires will have major impact on the fields of renewable energy, new (quasi) particles and quantum information processing. Recently, the first signatures of Majorana fermions have been observed in our InSb nanowires. With the proposed nanowire networks the special properties of this recently discovered particle can be tested for the first time."
Summary
"Nanowires are a powerful and versatile platform for a broad range of applications. Among all semiconductors, the heavy-elements materials exhibit the highest electron mobilities, strongest spin-orbit coupling and best thermoelectric properties. Nonetheless, heavy-element nanowires have been unexplored. With this proposal we unite the unique advantages of design freedom of nanowires with the special properties of heavy-element semiconductors. We specifically reveal the potential of heavy-element nanowires in the areas of thermoelectrics, and topological insulators. Using our strong track record in this area, we will pioneer the synthesis of this new class of materials and study their intrinsic materials properties. Starting point are nanowires of InSb and PbTe grown using the vapor-liquid-solid mechanism. Our aims are 1) to obtain highest-possible electron mobilities for these bottom-up fabricated materials by investigating new materials combinations of different semiconductor classes to effectively passivate the nanowire surface and we will eliminate impurities; 2) to investigate and optimize thermoelectric properties by developing advanced superlattice and core/shell nanowire structures where electronic and phononic transport is decoupled; and 3) to fabricate high-quality planar nanowire networks, which enable four-point electronic transport measurements and allow precisely determining carrier concentration and mobility. Besides the fundamentally interesting materials science, the heavy-element nanowires will have major impact on the fields of renewable energy, new (quasi) particles and quantum information processing. Recently, the first signatures of Majorana fermions have been observed in our InSb nanowires. With the proposed nanowire networks the special properties of this recently discovered particle can be tested for the first time."
Max ERC Funding
2 698 447 €
Duration
Start date: 2014-09-01, End date: 2019-08-31
