document

Deep Reinforcement Learning for Orchestrating Cost-Aware Reconfigurations of vRANs

Murti, Fahri Wisnu (author), Ali, Samad (author), Iosifidis, G. (author), Latva-aho, Matti (author)

Virtualized Radio Access Networks (vRANs) are fully configurable and can be implemented at a low cost over commodity platforms to enable network management flexibility. In this paper, a novel vRAN reconfiguration problem is formulated to jointly reconfigure the functional splits of the base stations (BSs), locations of the virtualized central...

journal article 2024

document

Control-relevant neural networks for feedforward control with preview: Applied to an industrial flatbed printer

Aarnoudse, Leontine (author), Kon, Johan (author), Ohnishi, Wataru (author), Poot, Maurice (author), Tacx, Paul (author), Strijbosch, Nard (author), Oomen, T.A.E. (author)

The performance of feedforward control depends strongly on its ability to compensate for reproducible disturbances. The aim of this paper is to develop a systematic framework for artificial neural networks (ANN) for feedforward control. The method involves three aspects: a new criterion that emphasizes the closed-loop control objective,...

journal article 2024

document

Neural inverse procedural modeling of knitting yarns from images

Trunz, Elena (author), Klein, Jonathan (author), Müller, Jan (author), Bode, Lukas (author), Sarlette, Ralf (author), Weinmann, M. (author), Klein, Reinhard (author)

We investigate the capabilities of neural inverse procedural modeling to infer high-quality procedural yarn models with fiber-level details from single images of depicted yarn samples. While directly inferring all parameters of the underlying yarn model based on a single neural network may seem an intuitive choice, we show that the complexity...

journal article 2024

document

Approximate dynamic programming for constrained linear systems: A piecewise quadratic approximation approach

He, K. (author), Shi, S. (author), van den Boom, A.J.J. (author), De Schutter, B.H.K. (author)

Approximate dynamic programming (ADP) faces challenges in dealing with constraints in control problems. Model predictive control (MPC) is, in comparison, well-known for its accommodation of constraints and stability guarantees, although its computation is sometimes prohibitive. This paper introduces an approach combining the two methodologies...

journal article 2024

document

Reliable and Energy-efficient Diabetic Retinopathy Screening using Memristor-based Neural Networks

Diware, S.S. (author), Chilakala, Koteswararao (author), Joshi, Rajiv V. (author), Hamdioui, S. (author), Bishnoi, R.K. (author)

Diabetic retinopathy (DR) is a leading cause of permanent vision loss worldwide. It refers to irreversible retinal damage caused due to elevated glucose levels and blood pressure. Regular screening for DR can facilitate its early detection and timely treatment. Neural network-based DR classifiers can be leveraged to achieve such screening in...

journal article 2024

document

Adaptive Prescribed Performance Asymptotic Tracking for High-Order Odd-Rational-Power Nonlinear Systems

Lv, Maolong (author), De Schutter, B.H.K. (author), Cao, Jinde (author), Baldi, S. (author)

Practical tracking results have been reported in the literature for high-order odd-rational-power nonlinear dynamics (a chain of integrators whose power is the ratio of odd integers). Asymptotic tracking remains an open problem for such dynamics. This note gives a positive answer to this problem in the framework of prescribed performance...

journal article 2023

document

Training Generative Adversarial Networks via Stochastic Nash Games

Franci, B. (author), Grammatico, S. (author)

Generative adversarial networks (GANs) are a class of generative models with two antagonistic neural networks: a generator and a discriminator. These two neural networks compete against each other through an adversarial process that can be modeled as a stochastic Nash equilibrium problem. Since the associated training process is challenging,...

journal article 2023

document

DeepPick: A Deep Learning Approach to Unveil Outstanding Users Ranking with Public Attainable Features

Li, Wanda (author), Xu, Zhiwei (author), Sun, Yi (author), Gong, Qingyuan (author), Chen, Y. (author), Ding, Aaron Yi (author), Wang, Xin (author), Hui, Pan (author)

Outstanding users (OUs) denote the influential, 'core' or 'bridge' users in online social networks. How to accurately detect and rank them is an important problem for third-party online service providers and researchers. Conventional efforts, ranging from early graph-based algorithms to recent machine learning-based approaches, typically rely on...

journal article 2023

document

Unrolling of Simplicial ElasticNet for Edge Flow Signal Reconstruction

Liu, Chengen (author), Leus, G.J.T. (author), Isufi, E. (author)

The edge flow reconstruction task consists of retreiving edge flow signals from corrupted or incomplete measurements. This is typically solved by a regularized optimization problem on higher-order networks such as simplicial complexes and the corresponding regularizers are chosen based on prior knowledge. Tailoring this prior to the setting...

journal article 2023

document

Symbolic knowledge extraction for explainable nutritional recommenders

Magnini, Matteo (author), Ciatto, Giovanni (author), Cantürk, Furkan (author), Aydoğan, Reyhan (author), Omicini, Andrea (author)

Background and objective:This paper focuses on nutritional recommendation systems (RS), i.e. AI-powered automatic systems providing users with suggestions about what to eat to pursue their weight/body shape goals. A trade-off among (potentially) conflictual requirements must be taken into account when designing these kinds of systems, there...

journal article 2023

document

Backdoor Pony: Evaluating backdoor attacks and defenses in different domains

Mercier, Arthur (author), Smolin, Nikita (author), Sihlovec, Oliver (author), Koffas, S. (author), Picek, S. (author)

Outsourced training and crowdsourced datasets lead to a new threat for deep learning models: the backdoor attack. In this attack, the adversary inserts a secret functionality in a model, activated through malicious inputs. Backdoor attacks represent an active research area due to diverse settings where they represent a real threat. Still,...

journal article 2023

document

Into the unknown: active monitoring of neural networks (extended version)

Kueffner, Konstantin (author), Lukina, A. (author), Schilling, Christian (author), Henzinger, Thomas A. (author)

Neural-network classifiers achieve high accuracy when predicting the class of an input that they were trained to identify. Maintaining this accuracy in dynamic environments, where inputs frequently fall outside the fixed set of initially known classes, remains a challenge. We consider the problem of monitoring the classification decisions of...

journal article 2023

document

Damage surrogate models for real-time flooding risk assessment of passenger ships

Mauro, F. (author), Conti, Fabien (author), Vassalos, Dracos (author)

Real-time assessment of flooding risk associated with the collision between two ships, requires a fast estimation of damage dimensions and associated survivability. The state-of-the-art frameworks for risk assessment on passenger ships do not consider a direct evaluation of damages through crash simulations but refer to probabilistic...

journal article 2023

document

Introducing CNN-LSTM network adaptations to improve remaining useful life prediction of complex systems

Borst, N.G. (author), Verhagen, W.J.C. (author)

Prognostics and Health Management (PHM) models aim to estimate remaining useful life (RUL) of complex systems, enabling lower maintenance costs and increased availability. A substantial body of work considers the development and testing of new models using the NASA C-MAPSS dataset as a benchmark. In recent work, the use of ensemble methods...

journal article 2023

document

Robust Lane Detection Through Self Pre-Training With Masked Sequential Autoencoders and Fine-Tuning With Customized PolyLoss

Li, Ruohan (author), Dong, Y. (author)

Lane detection is crucial for vehicle localization which makes it the foundation for automated driving and many intelligent and advanced driving assistant systems. Available vision-based lane detection methods do not make full use of the valuable features and aggregate contextual information, especially the interrelationships between lane...

journal article 2023

document

Safety Certification for Stochastic Systems via Neural Barrier Functions

Mathiesen, Frederik Baymler (author), Calvert, S.C. (author), Laurenti, L. (author)

Providing non-trivial certificates of safety for non-linear stochastic systems is an important open problem. One promising solution to address this problem is the use of barrier functions. Barrier functions are functions whose composition with the system forms a Martingale and enable the computation of the probability that the system stays...

journal article 2023

document

Reconstructing occluded Elevation Information in Terrain Maps with Self-supervised Learning

Stölzle, Maximilian (author), Miki, Takahiro (author), Gerdes, Levin (author), Azkarate, Martin (author), Hutter, Marco (author)

Accurate and complete terrain maps enhance the awareness of autonomous robots and enable safe and optimal path planning. Rocks and topography often create occlusions and lead to missing elevation information in the Digital Elevation Map (DEM). Currently, these occluded areas are either fully avoided during motion planning or the missing...

journal article 2022

document

Monte Carlo simulation of SDEs using GANs

van Rhijn, J. (author), Oosterlee, C.W. (author), Grzelak, L.A. (author), Liu, S. (author)

Generative adversarial networks (GANs) have shown promising results when applied on partial differential equations and financial time series generation. We investigate if GANs can also be used to approximate one-dimensional Ito ^ stochastic differential equations (SDEs). We propose a scheme that approximates the path-wise conditional...

journal article 2022

document

Adaptive neural sliding mode control for heterogeneous ship formation keeping considering uncertain dynamics and disturbances

You, Xu (author), Yan, Xinping (author), Liu, Jialun (author), Li, Shijie (author), Negenborn, R.R. (author)

This paper investigates the formation keeping problem of heterogeneous ships with underactuated inputs, uncertain dynamics, and environmental disturbances. The control objective is to make the heterogeneous followers keep the desired formation while tracking a leader. To solve the problem effectively, a novel virtual leader–follower formation...

journal article 2022

document

A PDE-free, neural network-based eddy viscosity model coupled with RANS equations

Xu, R. (author), Zhou, Xu Hui (author), Han, Jiequn (author), Dwight, R.P. (author), Xiao, Heng (author)

In fluid dynamics, constitutive models are often used to describe the unresolved turbulence and to close the Reynolds averaged Navier–Stokes (RANS) equations. Traditional PDE-based constitutive models are usually too rigid to calibrate with a large set of high-fidelity data. Moreover, commonly used turbulence models are based on the weak...

journal article 2022