arXiv subjects

Gard Spreemann

Publications and source records attributed to Gard Spreemann.

7 recordsLinked to original sources

arXiv2022

FORLORN: A Framework for Comparing Offline Methods and Reinforcement Learning for Optimization of RAN Parameters

The growing complexity and capacity demands for mobile networks necessitate innovative techniques for optimizing resource usage. Meanwhile, recent breakthroughs have brought Reinforcement Learning (RL) into the domain of continuous control of real-world systems. As a step towards RL-based network control, this paper introduces a new framework for benchmarking the performance of an RL agent in network environments simulated with ns-3. Within this framework, we demonstrate that an RL agent without domain-specific knowledge can learn how to efficiently adjust Radio Access Network (RAN) parameters to match offline optimization in static scenarios, while also adapting on the fly in dynamic scenarios, in order to improve the overall user experience. Our proposed framework may serve as a foundation for further work in developing workflows for designing RL-based RAN control algorithms.

cs.NI↗

arXiv2020

Simplicial Neural Networks

We present simplicial neural networks (SNNs), a generalization of graph neural networks to data that live on a class of topological spaces called simplicial complexes. These are natural multi-dimensional extensions of graphs that encode not only pairwise relationships but also higher-order interactions between vertices - allowing us to consider richer data, including vector fields and $n$-fold collaboration networks. We define an appropriate notion of convolution that we leverage to construct the desired convolutional neural networks. We test the SNNs on the task of imputing missing data on coauthorship complexes.

cs.LG↗

arXiv2019

A Notion of Harmonic Clustering in Simplicial Complexes

We outline a novel clustering scheme for simplicial complexes that produces clusters of simplices in a way that is sensitive to the homology of the complex. The method is inspired by, and can be seen as a higher-dimensional version of, graph spectral clustering. The algorithm involves only sparse eigenproblems, and is therefore computationally efficient. We believe that it has broad application as a way to extract features from simplicial complexes that often arise in topological data analysis.

cs.LG↗

arXiv2019

Same But Different: Distance Correlations Between Topological Summaries

Persistent homology allows us to create topological summaries of complex data. In order to analyse these statistically, we need to choose a topological summary and a relevant metric space in which this topological summary exists. While different summaries may contain the same information (as they come from the same persistence module), they can lead to different statistical conclusions since they lie in different metric spaces. The best choice of metric will often be application-specific. In this paper we discuss distance correlation, which is a non-parametric tool for comparing data sets that can lie in completely different metric spaces. In particular we calculate the distance correlation between different choices of topological summaries. We compare some different topological summaries for a variety of random models of underlying data via the distance correlation between the samples. We also give examples of performing distance correlation between topological summaries and other scalar measures of interest, such as a paired random variable or a parameter of the random model used to generate the underlying data. This article is meant to be expository in style, and will include the definitions of standard statistical quantities in order to be accessible to non-statisticians.

math.AT↗

arXiv2019

Topological exploration of artificial neuronal network dynamics

One of the paramount challenges in neuroscience is to understand the dynamics of individual neurons and how they give rise to network dynamics when interconnected. Historically, researchers have resorted to graph theory, statistics, and statistical mechanics to describe the spatiotemporal structure of such network dynamics. Our novel approach employs tools from algebraic topology to characterize the global properties of network structure and dynamics. We propose a method based on persistent homology to automatically classify network dynamics using topological features of spaces built from various spike-train distances. We investigate the efficacy of our method by simulating activity in three small artificial neural networks with different sets of parameters, giving rise to dynamics that can be classified into four regimes. We then compute three measures of spike train similarity and use persistent homology to extract topological features that are fundamentally different from those used in traditional methods. Our results show that a machine learning classifier trained on these features can accurately predict the regime of the network it was trained on and also generalize to other networks that were not presented during training. Moreover, we demonstrate that using features extracted from multiple spike-train distances systematically improves the performance of our method.

q-bio.NC↗

arXiv2015

Using persistent homology to reveal hidden information in neural data

We propose a method, based on persistent homology, to uncover topological properties of a priori unknown covariates of neuron activity. Our input data consist of spike train measurements of a set of neurons of interest, a candidate list of the known stimuli that govern neuron activity, and the corresponding state of the animal throughout the experiment performed. Using a generalized linear model for neuron activity and simple assumptions on the effects of the external stimuli, we infer away any contribution to the observed spike trains by the candidate stimuli. Persistent homology then reveals useful information about any further, unknown, covariates.

q-bio.NC↗

arXiv2014

Approximating Persistent Homology in Euclidean Space Through Collapses

The Čech complex is one of the most widely used tools in applied algebraic topology. Unfortunately, due to the inclusive nature of the Čech filtration, the number of simplices grows exponentially in the number of input points. A practical consequence is that computations may have to terminate at smaller scales than what the application calls for. In this paper we propose two methods to approximate the Čech persistence module. Both are constructed on the level of spaces, i.e. as sequences of simplicial complexes induced by nerves. We also show how the bottleneck distance between such persistence modules can be understood by how tightly they are sandwiched on the level of spaces. In turn, this implies the correctness of our approximation methods. Finally, we implement our methods and apply them to some example point clouds in Euclidean space.

math.AT↗