Publications related to A statistical physics perspective of complex networks

D-Cliques: Compensating for Data Heterogeneity with Topology in Decentralized Federated Learning

The convergence speed of machine learning models trained with Federated Learning is significantly affected by non-independent and identically distributed (non-IID) data partitions, even more so in a fully decentralized setting without a central server. In ...

2022

Information Retrieval Under Network Uncertainty: Robust Internet Ranking

Ralf Seifert, Anna Timonina-Farkas

Internet ranking algorithms play a crucial role in information technologies and numerical analysis due to their efficiency in high dimensions and wide range of possible applications, including scientometrics and systemic risk in finance (SinkRank, DebtRank ...

INFORMS2022

An automatized signal timing detection at signalized intersections with detailed vehicle trajectory data

Nikolaos Geroliminis, Emmanouil Barmpounakis, Jasso Espadaler Clapés

Signalized intersections are a fundamental part of urban networks. Their understanding is crucial to identify congestion patterns, queues, delays, and safety issues in local and network level. In this work, we analyze multimodal vehicle trajectories and pr ...

2022

Efficient Max-Pressure Traffic Management for Large-Scale Congested Urban Networks

Nikolaos Geroliminis, Dimitrios Tsitsokas, Anastasios Kouvelas

Traffic responsive signal control systems bear high potential in reducing delays in congested networks due to their ability of dynamically adjusting right-of-way assignment among conflicting movements, based on real-time traffic measurements. In this work, ...

2022

Supplementary datasets for "ARBRE: Computational resource to predict pathways towards industrially important aromatic compounds"

Vassily Hatzimanikatis, Homa Mohammadi Peyhani, Anastasia Sveshnikova

Supplementary datasets accompanying the manuscript "ARBRE: Computational resource to predict pathways towards industrially important aromatic compounds" published in the Metabolic Engineering Journal (https://doi.org/10.1016/j.ymben.2022.03.013). In line w ...

EPFL Infoscience2022

True scale-free networks hidden by finite size effects

Andrea Rinaldo, Samir Suweis, Amos Maritan

We analyze about 200 naturally occurring networks with distinct dynamical origins to formally test whether the commonly assumed hypothesis of an underlying scale-free structure is generally viable. This has recently been questioned on the basis of statisti ...

NATL ACAD SCIENCES2021

The influence of trip length distribution on urban traffic in network-level models

Mikhail Murashkin

Modeling urban traffic on the level of network is a wide research area oriented to the development of ITS. In this thesis properties of models based on MFD (Macroscopic Fundamental Diagram) are studied. The idea behind MFD is to say that the state of the t ...

EPFL2021

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

NICEpath: Finding metabolic pathways in large networks through atom-conserving substrate-product pairs

Vassily Hatzimanikatis, Jasmin Maria Hafner

Motivation: Finding biosynthetic pathways is essential for metabolic engineering of organisms to produce chemicals, biodegradation prediction of pollutants and drugs, and for the elucidation of bioproduction pathways of secondary metabolites. A key step in ...

OXFORD UNIV PRESS2021

Quasi-Global Momentum: Accelerating Decentralized Deep Learning on Heterogeneous Data

Martin Jaggi, Sebastian Urban Stich, Tao Lin, Sai Praneeth Reddy Karimireddy

Decentralized training of deep learning models is a key element for enabling data privacy and on-device learning over networks. In realistic learning scenarios, the presence of heterogeneity across different clients' local datasets poses an optimization ch ...

JMLR-JOURNAL MACHINE LEARNING RESEARCH2021

A statistical physics perspective of complex networks

Graph Chatbot

Chat with Graph Search

D-Cliques: Compensating for Data Heterogeneity with Topology in Decentralized Federated Learning

Information Retrieval Under Network Uncertainty: Robust Internet Ranking

An automatized signal timing detection at signalized intersections with detailed vehicle trajectory data

Efficient Max-Pressure Traffic Management for Large-Scale Congested Urban Networks

Supplementary datasets for "ARBRE: Computational resource to predict pathways towards industrially important aromatic compounds"

True scale-free networks hidden by finite size effects

The influence of trip length distribution on urban traffic in network-level models

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

NICEpath: Finding metabolic pathways in large networks through atom-conserving substrate-product pairs

Quasi-Global Momentum: Accelerating Decentralized Deep Learning on Heterogeneous Data

Information Retrieval Under Network Uncertainty: Robust Internet Ranking

True scale-free networks hidden by finite size effects

The influence of trip length distribution on urban traffic in network-level models

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

Quasi-Global Momentum: Accelerating Decentralized Deep Learning on Heterogeneous Data

NICEpath: Finding metabolic pathways in large networks through atom-conserving substrate-product pairs

Supplementary datasets for "ARBRE: Computational resource to predict pathways towards industrially important aromatic compounds"

Efficient Max-Pressure Traffic Management for Large-Scale Congested Urban Networks

D-Cliques: Compensating for Data Heterogeneity with Topology in Decentralized Federated Learning

An automatized signal timing detection at signalized intersections with detailed vehicle trajectory data