Publications by Andreas Loukas

Chat with Graph Search

Ask any question about EPFL courses, lectures, exercises, research, news, etc. or try the example questions below.

DISCLAIMER: The Graph Chatbot is not programmed to provide explicit or categorical answers to your questions. Rather, it transforms your questions into API requests that are distributed across the various IT services officially administered by EPFL. Its purpose is solely to collect and recommend relevant references to content that you can explore to help you answer your questions.

An emerging way to deal with high-dimensional non-euclidean data is to assume that the underlying structure can be captured by a graph. Recently, ideas have begun to emerge related to the analysis of time-varying graph signals. This work aims to elevate th ...

Graph filters play a key role in processing the graph spectra of signals supported on the vertices of a graph. However, despite their widespread use, graph filters have been analyzed only in the deterministic setting, ignoring the impact of stochasticity i ...

This paper initiates the study of the classic balanced graph partitioning problem from an online perspective: Given an arbitrary sequence of pairwise communication requests between n nodes, with patterns that may change over time, the objective is to servi ...

Integrating functions on discrete domains into neural networks is key to developing their capability to reason about discrete objects. But, discrete domains are (I) not naturally amenable to gradient-based optimization, and (II) incompatible with deep lear ...

Proteins are typically represented by discrete atomic coordinates providing an accessible framework to describe different conformations. However, in some fields proteins are more accurately represented as near-continuous surfaces, as these are imprinted wi ...

A hallmark of graph neural networks is their ability to distinguish the isomorphism class of their inputs. This study derives hardness results for the classification variant of graph isomorphism in the message-passing model (MPNN). MPNN encompasses the maj ...

Graph-based methods for signal processing have shown promise for the analysis of data exhibiting irregular structure, such as those found in social, transportation, and sensor networks. Yet, though these systems are often dynamic, state-of-the-art methods ...

In graph coarsening, one aims to produce a coarse graph of reduced size while preserving important graph properties. However, as there is no consensus on which specific graph properties should be preserved by coarse graphs, measuring the differences betwee ...

How many samples are sufficient to guarantee that the eigenvectors and eigenvalues of the sample covariance matrix are close to those of the actual covariance matrix? For a wide family of distributions, including distributions with finite second moment and ...

Can we use machine learning to compress graph data? The absence of ordering in graphs poses a significant challenge to conventional compression algorithms, limiting their attainable gains as well as their ability to discover relevant patterns. On the other ...

We approach the graph generation problem from a spectral perspective by first generating the dominant parts of the graph Laplacian spectrum and then building a graph matching these eigenvalues and eigenvectors. Spectral conditioning allows for direct model ...

We consider the problem of inferring the hidden structure of high-dimensional dynamic systems from the perspective of graph learning. In particular, we aim at capturing the dynamic relationships between nodes by a sequence of graphs. Our approach is motiva ...

How does coarsening affect the spectrum of a general graph? We provide conditions such that the principal eigenvalues and eigenspaces of a coarsened and original graph Laplacian matrices are close. The achieved approximation is shown to depend on standard ...

Attention-based architectures have become ubiquitous in machine learning. Yet, our understanding of the reasons for their effectiveness remains limited. This work proposes a new way to understand self-attention networks: we show that their output can be de ...

We present a novel implementation strategy for distributed autoregressive moving average (ARMA) graph filters. Differently from the state of the art implementation, the proposed approach has the following benefits: (i) the designed filter coefficients come ...

Spectral clustering is a widely studied problem, yet its complexity is prohibitive for dynamic graphs of even modest size. We claim that it is possible to get information from past cluster assignments to expedite computation. Our approach builds on a recen ...

De novo protein design explores uncharted sequence and structure space to generate novel proteins not sampled by evolution. A main challenge in de novo design involves crafting “designable” structural templates to guide the sequence searches toward adoptin ...

This paper studies the expressive power of graph neural networks falling within the message-passing framework (GNNmp). Two results are presented. First, GNNmp are shown to be Turing universal under sufficient conditions on their depth, width, node attribut ...

In this paper, we present a graph partitioning algorithm to partition graphs with trillions of edges. To achieve such scale, our solution leverages the vertex-centric Pregel abstraction provided by Giraph, a system for large-scale graph analytics. We desig ...

Andreas Loukas

Chat with Graph Search

A Time-Vertex Signal Processing Framework

Filtering Random Graph Processes Over Random Time-Varying Graphs

Dynamic Balanced Graph Partitioning

Neural Set Function Extensions: Learning with Discrete Functions in High Dimensions

RosettaSurf-A surface-centric computational design approach

How hard is to distinguish graphs with graph neural networks?

Towards Stationary Time-Vertex Signal Processing

Graph Coarsening with Preserved Spectral Properties

How close are the eigenvectors and eigenvalues of the sample and actual covariance matrices?

Partition and Code: learning how to compress graphs

SPECTRE: Spectral Conditioning Helps to Overcome the Expressivity Limits of One-shot Graph Generators

Learning time varying graphs

Spectrally approximating large graphs with smaller graphs

Attention is not all you need: pure attention loses rank doubly exponentially with depth

Autoregressive moving average graph filters a stable distributed implementation

Fast Approximate Spectral Clustering for Dynamic Networks

Exploring “dark-matter” protein folds using deep learning

What graph neural networks cannot learn: depth vs width

Spinner: Scalable Graph Partitioning in the Cloud

Forecasting Time Series With VARMA Recursions on Graphs