Publications associées à Landscape-scale conservation

Vers un projet territorial du vivant L'outil des réseaux écologiques comme projet urbanistique de revitalisation du paysage agricole

The aim of my thesis is to provide urban architects and planners with working theoretical elements territorial project of the living. The living and the socio-ecological transition are key concepts in the nature and people paradigm, which asserts a human-n ...

EPFL2024

Quartiers rhodaniens en transition

Sara Sonia Formery Regazzoni

Over the course of history, the relationship between cities and their waters has shown different gradients of interweaving, marked by cycles of bonding and distancing. Following a period of complete neglect of urban watercourses, the versatile, multifacete ...

EPFL2024

The effect of smooth parametrizations on nonconvex optimization landscapes

Nicolas Boumal

We develop new tools to study landscapes in nonconvex optimization. Given one optimization problem, we pair it with another by smoothly parametrizing the domain. This is either for practical purposes (e.g., to use smooth optimization algorithms with good g ...

Springer Heidelberg2024

Genomics and conservation: Guidance from training to analyses and applications

Annie Sandrine Guillaume

Environmental change is intensifying the biodiversity crisis and threatening species across the tree of life. Conservation genomics can help inform conservation actions and slow biodiversity loss. However, more training, appropriate use of novel genomic me ...

Wiley2023

Digitised Historical Newspapers: A Changing Research Landscape (Introduction)

Maud Ehrmann, Frédéric Clavert

The application of digital technologies to newspaper archives is transforming the way historians engage with these sources. The digital evolution not only affects how scholars access historical newspapers, but also, increasingly, how they search, explore a ...

De Gruyter Oldenbourg2022

Conduit, Patio, Waste: Mapping Environmental Relations in Bairro da Malagueira

Bárbara Maçães E Costa

The ongoing ecological crisis is reframing the common binary opposition between architecture and nature and provoking us to ask what kinds of buildings we collectively want to create for the ecosystems that we inhabit. The environment is not a neutral back ...

EPFL2021

On the relation between landscape beauty and land cover: A case study in the U.K. at Sentinel-2 resolution with interpretable AI

Devis Tuia, Alex Hubertus Levering

The environment where we live and recreate can have a significant effect on our well-being. More beautiful landscapes have considerable benefits to both health and quality of life. When we chose where to live or our next holiday destination, we do so accor ...

2021

Stochasticity helps to navigate rough landscapes: comparing gradient-descent-based algorithms in the phase retrieval problem

Lenka Zdeborová, Francesca Mignacco

In this paper we investigate how gradient-based algorithms such as gradient descent (GD), (multi-pass) stochastic GD, its persistent variant, and the Langevin algorithm navigate non-convex loss-landscapes and which of them is able to reach the best general ...

2021

Geometry of the Loss Landscape in Overparameterized Neural Networks: Symmetries and Invariances

Wulfram Gerstner, Clément Hongler, Johanni Michael Brea, Francesco Spadaro, Berfin Simsek, Arthur Jacot

We study how permutation symmetries in overparameterized multi-layer neural networks generate `symmetry-induced' critical points. Assuming a network with

L

layers of minimal widths

r_1^*, \ldots, r_{L-1}^*

reaches a zero-loss minimum at $ r_1^*! \c ...

2021

On the Loss Landscape of Adversarial Training: Identifying Challenges and How to Overcome Them

Sabine Süsstrunk, Mathieu Salzmann, Tao Lin, Chen Liu

We analyze the influence of adversarial training on the loss landscape of machine learning models. To this end, we first provide analytical studies of the properties of adversarial loss functions under different adversarial budgets. We then demonstrate tha ...

2021

Spatiotemporal patterns of urbanization in three Swiss urban agglomerations: insights from landscape metrics, growth modes and fractal analysis

Jérôme Chenal, Martí Bosch Padrós, Rémi Jaligot

Context Urbanization is the most important form of landscape change and is increasingly affecting biodiversity and ecosystem functions. Understanding how landscape patterns change in space and time is central to the evaluation of the environmental impacts ...

SPRINGER2020

Beyond the patch: on landscape-explicit metapopulation dynamics

Jonathan Giezendanner

Climate change threatens biodiversity and species distribution all over the world at unprecedented rates. Human induced changes to landscape structure and habitat are redefining the relation between species and their environment. Understanding, characteriz ...

EPFL2020

Towards sustainable transitions of peri-urban neighborhoods of detached houses

Judith Elisa Drouilles, Emmanuel Rey

Peri-urban neighborhoods of detached houses are facing countless challenges related to the building stock’s low energy efficiency, to the current demographic transition reducing household size and increasing the proportion of elderlies, to their distant lo ...

2020

Edge influence on understorey plant communities depends on forest management

Matteo Tolosano

Questions Does the influence of forest edges on plant species richness and composition depend on forest management? Do forest specialists and generalists show contrasting patterns? ...

WILEY2020

Dynamical mean-field theory for stochastic gradient descent in Gaussian mixture classification

Florent Gérard Krzakala, Lenka Zdeborová, Francesca Mignacco

We analyze in a closed form the learning dynamics of stochastic gradient descent (SGD) for a single layer neural network classifying a high-dimensional Gaussian mixture where each cluster is assigned one of two labels. This problem provides a prototype of ...

Curran Associates, Inc.2020

PyLandStats: An open-source Pythonic library to compute landscape metrics

Martí Bosch Padrós

Quantifying the spatial pattern of landscapes has become a common task of many studies in landscape ecology. Most of the existing software to compute landscape metrics is not well suited to be used in interactive environments such as Jupyter notebooks nor ...

2019

Evolving biodiversity patterns in changing river networks

Andrea Rinaldo, Enrico Bertuzzo

Biodiversity patterns are governed by landscape structure and dispersal strategies of residing organisms. Landscape, however, changes, and dispersal strategies evolve with it. It is unclear how these biological and geomorphological changes interplay to aff ...

2019

Comparing dynamics: deep neural networks versus glassy systems

Matthieu Wyart, Mario Geiger, Marco Baity-Jesi, Stefano Spigler, Levent Dogus Sagun

We analyze numerically the training dynamics of deep neural networks (DNN) by using methods developed in statistical physics of glassy systems. The two main issues we address are (1) the complexity of the loss landscape and of the dynamics within it, and ( ...

IOP PUBLISHING LTD2019

Microbial life in porous systems

David Scheidweiler

Microbial life in porous systems dominates the functioning of numerous ecosystems, ranging from stream sediments to soils. While these environments are characterized by structures that vary spatially over orders of magnitude, the traditional research focus ...

EPFL2019

Jamming transition as a paradigm to understand the loss landscape of deep neural networks

Matthieu Wyart, Mario Geiger, Marco Baity-Jesi, Stefano Spigler, Levent Dogus Sagun

Deep learning has been immensely successful at a variety of tasks, ranging from classification to artificial intelligence. Learning corresponds to fitting training data, which is implemented by descending a very high-dimensional loss function. Understandin ...

AMER PHYSICAL SOC2019