Venue: ENS Paris, 24 rue Lhomond 75005 Paris, room L357/L359.
Dates: 4-6 April 2018
Funding: The workshop is funded by the Marie Skłodowska-Curie project “Philosophical Foundations of Topological Explanations” (TOPEX), through People Programme (Marie Curie Actions) of the European Union's H2020-MSCA-IF-2015 Programme under REA grant agreement n° (703662). The workshop is also supported by the École normale supérieure Paris.
Workshop description
Over the last two decades, network based approaches to modelling complex systems and explanations based on them have become ubiquitous in many areas of science, such as in cognitive neuroscience in the network analysis in the Connectome project, implemented in two Big Science projects namely the BRAIN Initiative (BRAIN stands for “Brain Research through Advancing Innovative Neurotechnologies”) in the US, and the Human Brain Project in the EU, in ecology, e.g. stability, resilience and robustness of ecological communities, trophic webs in biology, e.g. protein and metabolic interaction networks, in medicine, e.g. gene networks for disease.
Even though it is encouraging that the interest in of this approach grows very rapidly, both in philosophy and in the sciences, especially in the last couple of years, some of the most important conceptual and methodological issues, still require a programmatic foundation. The richness and complexity of connectivity patterns in real world systems as well as the specific questions that seek explanation in various sciences, are reflected in the plurality and diversity of modelling approaches and explanatory patterns in different areas of science.
But the question is: can there be a generalised account of network explanation that can be applied across sciences?
A need for the generalised account and programmatic and foundational treatment of key conceptual and methodological issues of network approaches in sciences, especially in neuroscience, is particularly pressing because network models in neuroscience mainly just show how already known network structures from graph theory map onto the brain networks. However, we need a method that can allow for finer-grain understanding of brain functions and probing into the actual dependencies between the brain network structures and brain function. This point can be further refined and applied to other areas of science that utilize network approach in terms of levels in network organisation, e.g. until very recently the notion of small-world topology was considered as crucial aspect for understanding the efficiency of the brain organisation in signal processing. However, given the multitudes of ways in which the arrangements of topological properties can be used to describe the small-world topology, and the ways in which these arrangements provide different patterns of dependencies and constraints between topological structure and the dynamical features of the system, it rapidly becomes unclear what is the explanatory role of higher level global network properties such as small-worldiness.
In ecology, genetics and theory of complex systems the similar questions arise, i.e. does spatial embedding require different ways to describe networks, what is the role of timescale in the evolution of networks and how to capture it, and many other.
Philosophical interest in these issues stems from the emerging debates on mathematical and non-causal explanations. For example, it is widely regarded that a good explanation has to be asymmetric (if A explains B, then B cannot explain A, otherwise the explanation is circular). In causal explanations, the explanatory asymmetry simply follows the causal asymmetry, i.e. the causes explain their effects, and not the other way around. However, if we assume that non-causal explanations (explanations that appeal to facts or properties that are not considered causes, e.g. a parent can’t divide 23 strawberries so that each of their children gets an equal number of strawberries, because the number 23 is not divisible by 3 to give an even number) are also asymmetric, whence does their asymmetry come from, if it isn't from some causal asymmetry? In virtue of what something is representational, or has a representational role in explanation and are there any norms for distinguishing those roles? Even though the non-causal explanations don't hinge on contingent empirical facts, but instead describe mathematical necessities, they should not be seen as being completely divorced from reality, and hence false. So, what kind of empirical facts they do have to take into account, in order to avoid being divorced from reality? Is there a norm for sorting a distinctly network target phenomenon that is to be explained?
This highly interdisciplinary workshop will include some of the most renown scientists and philosophers who are working on theories of explanations and on network approaches to levels, hierarchies and asymmetries. This workshop and the subsequent publication of its proceedings will provide a unique opportunity for the direct impact of philosophy on science (through the development of conceptual and methodological norms for understanding the levels and asymmetries) as well as a framework for the direct influence of science to philosophy (through modelling of the conceptual and methodological norms based on actual empirical research).
Participants:
Dates: 4-6 April 2018
Funding: The workshop is funded by the Marie Skłodowska-Curie project “Philosophical Foundations of Topological Explanations” (TOPEX), through People Programme (Marie Curie Actions) of the European Union's H2020-MSCA-IF-2015 Programme under REA grant agreement n° (703662). The workshop is also supported by the École normale supérieure Paris.
Workshop description
Over the last two decades, network based approaches to modelling complex systems and explanations based on them have become ubiquitous in many areas of science, such as in cognitive neuroscience in the network analysis in the Connectome project, implemented in two Big Science projects namely the BRAIN Initiative (BRAIN stands for “Brain Research through Advancing Innovative Neurotechnologies”) in the US, and the Human Brain Project in the EU, in ecology, e.g. stability, resilience and robustness of ecological communities, trophic webs in biology, e.g. protein and metabolic interaction networks, in medicine, e.g. gene networks for disease.
Even though it is encouraging that the interest in of this approach grows very rapidly, both in philosophy and in the sciences, especially in the last couple of years, some of the most important conceptual and methodological issues, still require a programmatic foundation. The richness and complexity of connectivity patterns in real world systems as well as the specific questions that seek explanation in various sciences, are reflected in the plurality and diversity of modelling approaches and explanatory patterns in different areas of science.
But the question is: can there be a generalised account of network explanation that can be applied across sciences?
A need for the generalised account and programmatic and foundational treatment of key conceptual and methodological issues of network approaches in sciences, especially in neuroscience, is particularly pressing because network models in neuroscience mainly just show how already known network structures from graph theory map onto the brain networks. However, we need a method that can allow for finer-grain understanding of brain functions and probing into the actual dependencies between the brain network structures and brain function. This point can be further refined and applied to other areas of science that utilize network approach in terms of levels in network organisation, e.g. until very recently the notion of small-world topology was considered as crucial aspect for understanding the efficiency of the brain organisation in signal processing. However, given the multitudes of ways in which the arrangements of topological properties can be used to describe the small-world topology, and the ways in which these arrangements provide different patterns of dependencies and constraints between topological structure and the dynamical features of the system, it rapidly becomes unclear what is the explanatory role of higher level global network properties such as small-worldiness.
In ecology, genetics and theory of complex systems the similar questions arise, i.e. does spatial embedding require different ways to describe networks, what is the role of timescale in the evolution of networks and how to capture it, and many other.
Philosophical interest in these issues stems from the emerging debates on mathematical and non-causal explanations. For example, it is widely regarded that a good explanation has to be asymmetric (if A explains B, then B cannot explain A, otherwise the explanation is circular). In causal explanations, the explanatory asymmetry simply follows the causal asymmetry, i.e. the causes explain their effects, and not the other way around. However, if we assume that non-causal explanations (explanations that appeal to facts or properties that are not considered causes, e.g. a parent can’t divide 23 strawberries so that each of their children gets an equal number of strawberries, because the number 23 is not divisible by 3 to give an even number) are also asymmetric, whence does their asymmetry come from, if it isn't from some causal asymmetry? In virtue of what something is representational, or has a representational role in explanation and are there any norms for distinguishing those roles? Even though the non-causal explanations don't hinge on contingent empirical facts, but instead describe mathematical necessities, they should not be seen as being completely divorced from reality, and hence false. So, what kind of empirical facts they do have to take into account, in order to avoid being divorced from reality? Is there a norm for sorting a distinctly network target phenomenon that is to be explained?
This highly interdisciplinary workshop will include some of the most renown scientists and philosophers who are working on theories of explanations and on network approaches to levels, hierarchies and asymmetries. This workshop and the subsequent publication of its proceedings will provide a unique opportunity for the direct impact of philosophy on science (through the development of conceptual and methodological norms for understanding the levels and asymmetries) as well as a framework for the direct influence of science to philosophy (through modelling of the conceptual and methodological norms based on actual empirical research).
Participants:
- Marc Tittgemeyer (Head of Research Group for Translational Neurocircuitry, Max Planck Institute for Metabolism Research, Cologne, Germany).
- Daniel Kostic (CNRS/ IHPST/Université Paris 1 Panthéon-Sorbonne).
- Lina Jansson (Department of Philosophy, University of Nottingham).
- Maria Serban (Department of Media, Cognition and Communication, University of Copenhagen).
- David Chavalarias (Director of the Complex Systems Institute of Paris Ile-de-France, Centre d’Analyses de Mathématiques Sociales, Ecole des Hautes Etudes en Sciences Sociales).
- Elisa Thebault (CNRS/ UPMC - iEES Paris).
- Nathalie Niquil (UMR BOREA, Normandie Université, UNICAEN).
- Denis Forest (CNRS/ IHPST/Université Paris 1 Panthéon-Sorbonne).
- Sophie Achard (CNRS, GIPSA-lab, Grenoble, France).
- Philippe Huneman (CNRS/ IHPST/Université Paris 1 Panthéon-Sorbonne).
- Eric Bapteste (Université Pierre et Marie Curie, UMR 7138).
- Claus Hilgetag (Director of the Institute of Computational Neuroscience in Hamburg, Germany).
- Alex Goulas (Institute of Computational Neuroscience in Hamburg, Germany).
PROGRAM
04 April 2018 – ENS Paris, 24 rue Lhomond 75005 Paris, room L357/L359
10:00-10:15 – Daniel Kostic: Welcome and opening of the workshop.
10:15-11:00 – Marc Tittgemeyer: Exploring the hierarchical organization of brain networks.
11:00-11:45 – Daniel Kostic: Non-causal Asymmetries in Topological Explanations.
11:45-12:00 – Coffee & Tea pause
12:00-12:45 –Lina Jansson: Directionality in Non-Causal Models.
13:00-15:00 – Lunch break
15:00-15:45 – Maria Serban: Hierarchical modularity in biological networks.
15:45-16:30 – David Chavalarias: Multi-scale dynamics reconstruction of socio-semantic networks.
16:30-16:45 – Coffee & Tea pause.
17:00-18:00 – 1st day panel discussion chaired by Nathalie Niquil.
19:00 – Dinner.
05 April 2018 – ENS Paris, 24 rue Lhomond 75005 Paris, room L357/L359
10:00-10:45 – Elisa Thebault: Structure and stability of networks with mutualistic and antagonistic interactions.
10:45-11:30 – Nathalie Niquil: How theories of ecological networks and autocatalytic dynamics have diffused from ecology into multiple disciplines?
11:30-11:45 – Coffee & Tea pause
11:45-12:30 – Denis Forest: Connectomics, Why-questions and contrast-classes.
12:30-14:30 – Lunch break
14:30-15:15 – Sophie Achard: Assessing reliability of resting-state fMRI graph analysis: challenges in measuring brain connectivity networks alterations for clinical applications.
15:15-16:00 – Philippe Huneman: Realism, Genericity and topologies: a plea for the specficity of network explanations.
16:00-16:15 – Coffee & Tea pause
16:30-17:15 – 2nd day and general panel discussion chaired by Maria Serban and Denis Forest
19:30 –Official workshop dinner.
06 April 2018 – ENS Paris, 24 rue Lhomond 75005 Paris, room L357/L359
10:00-10:45 – Eric Bapteste: Informal insights on the use of networks in evolutionary Biology.
10:45-11:30 – Claus Hilgetag: “Hierarchy" in the organisation of brain networks.
11:30-11:45 – Coffee & Tea pause
11:45-12:30 – Alex Goulas: The wiring of the cerebral cortex from a first principles standpoint.
12:30-14:30 – Concluding remarks, closing the workshop, informal lunch and saying goodbye.
10:00-10:15 – Daniel Kostic: Welcome and opening of the workshop.
10:15-11:00 – Marc Tittgemeyer: Exploring the hierarchical organization of brain networks.
11:00-11:45 – Daniel Kostic: Non-causal Asymmetries in Topological Explanations.
11:45-12:00 – Coffee & Tea pause
12:00-12:45 –Lina Jansson: Directionality in Non-Causal Models.
13:00-15:00 – Lunch break
15:00-15:45 – Maria Serban: Hierarchical modularity in biological networks.
15:45-16:30 – David Chavalarias: Multi-scale dynamics reconstruction of socio-semantic networks.
16:30-16:45 – Coffee & Tea pause.
17:00-18:00 – 1st day panel discussion chaired by Nathalie Niquil.
19:00 – Dinner.
05 April 2018 – ENS Paris, 24 rue Lhomond 75005 Paris, room L357/L359
10:00-10:45 – Elisa Thebault: Structure and stability of networks with mutualistic and antagonistic interactions.
10:45-11:30 – Nathalie Niquil: How theories of ecological networks and autocatalytic dynamics have diffused from ecology into multiple disciplines?
11:30-11:45 – Coffee & Tea pause
11:45-12:30 – Denis Forest: Connectomics, Why-questions and contrast-classes.
12:30-14:30 – Lunch break
14:30-15:15 – Sophie Achard: Assessing reliability of resting-state fMRI graph analysis: challenges in measuring brain connectivity networks alterations for clinical applications.
15:15-16:00 – Philippe Huneman: Realism, Genericity and topologies: a plea for the specficity of network explanations.
16:00-16:15 – Coffee & Tea pause
16:30-17:15 – 2nd day and general panel discussion chaired by Maria Serban and Denis Forest
19:30 –Official workshop dinner.
06 April 2018 – ENS Paris, 24 rue Lhomond 75005 Paris, room L357/L359
10:00-10:45 – Eric Bapteste: Informal insights on the use of networks in evolutionary Biology.
10:45-11:30 – Claus Hilgetag: “Hierarchy" in the organisation of brain networks.
11:30-11:45 – Coffee & Tea pause
11:45-12:30 – Alex Goulas: The wiring of the cerebral cortex from a first principles standpoint.
12:30-14:30 – Concluding remarks, closing the workshop, informal lunch and saying goodbye.