document

Predictive Maintenance Planning Using Renewal Reward Processes and Probabilistic RUL Prognostics: Analyzing the Influence of Accuracy and Sharpness of Prognostics

Mitici, M.A. (author), de Pater, I.I. (author), Zeng, Zhiguo (author), Barros, Anne (author)

We pose the maintenance planning for systems using probabilistic Remaining Useful Life (RUL) prognostics as a renewal reward process. Data-driven probabilistic RUL prognostics are obtained using a Convolutional Neural Network with Monte Carlo dropout. The maintenance planning model is illustrated for aircraft turbofan engines. The results show...

conference paper 2023

document

Jumping Shift: A Logarithmic Quantization Method for Low-Power CNN Acceleration

Jiang, Longxing (author), Aledo Ortega, D. (author), van Leuken, T.G.R.M. (author)

Logarithmic quantization for Convolutional Neural Networks (CNN): a) fits well typical weights and activation distributions, and b) allows the replacement of the multiplication operation by a shift operation that can be implemented with fewer hardware resources. We propose a new quantization method named Jumping Log Quantization (JLQ). The key...

conference paper 2023

document

MNEMOSENE++: Scalable Multi-Tile Design with Enhanced Buffering and VGSOT-MRAM based Compute-in-Memory Crossbar Array

Escuin, Carlos (author), García-Redondo, Fernando (author), Zahedi, M.Z. (author), Ibáñez, Pablo (author), Monreal, Teresa (author), Viñals, Victor (author), Llabería, José María (author), Myers, James (author), Ryckaert, Julien (author), Biswas, Dwaipayan (author), Catthoor, Francky (author)

This paper optimizes the MNEMOSENE architecture, a compute-in-memory (CiM) tile design integrating computation and storage for increased efficiency. We identify and address bottlenecks in the Row Data (RD) buffer that cause losses in performance. Our proposed approach includes mitigating these buffering bottlenecks and extending MNEMOSENE’s...

conference paper 2023

document

Constructing Health Indicators for Systems with Few Failure Instances Using Unsupervised Learning

de Pater, I.I. (author), Mitici, M.A. (author)

Health indicators are crucial to assess the health of complex systems. In recent years, several studies have developed data-driven health indicators using supervised learning methods. However, due to preventive maintenance, there are often not enough failure instances to train a supervised learning model, i.e., the data is unlabelled with an...

conference paper 2023

document

Infrared thermal defect identification and reconstruction of artworks using a spatiotemporal deep neural network

Moradi, M. (author), Ghorbani, R. (author), Sfarra, Stefano (author), Tax, D.M.J. (author), Zarouchas, D. (author)

Assessment of cultural heritage assets is now extremely important all around the world. Non-destructive inspection is essential for preserving the integrity of the artworks while avoiding the loss of any precious materials that make it up. The use of Infrared Thermography (IRT) is an interesting concept since surface and subsurface faults can be...

conference paper 2022

document

Elderly Care - Human Activity Recognition Using Radar with an Open Dataset and Hybrid Maps

Zhang, Xinyu (author), Abbasi, Qammer H. (author), Fioranelli, F. (author), Romain, Olivier (author), Le Kernec, Julien (author)

Population ageing has become a severe problem worldwide. Human activity recognition (HAR) can play an important role to provide the elders with in-time healthcare. With the advantages of environmental insensitivity, contactless sensing and privacy protection, radar has been widely used for human activity detection. The micro-Doppler...

conference paper 2022

document

BackboneAnalysis: Structured Insights into Compute Platforms from CNN Inference Latency

Hafner, Frank M. (author), Zeller, Matthias (author), Schutera, Mark (author), Abhau, Jochen (author), Kooij, J.F.P. (author)

Customization of a convolutional neural network (CNN) to a specific compute platform involves finding an optimal pareto state between computational complexity of the CNN and resulting throughput in operations per second on the compute platform. However, existing inference performance benchmarks compare complete backbones that entail many...

conference paper 2022

document

Long-term behaviour recognition in videos with actor-focused region attention

Ballan, Luca (author), Strafforello, O. (author), Schutte, Klamer (author)

Long-Term activities involve humans performing complex, minutes-long actions. Differently than in traditional action recognition, complex activities are normally composed of a set of sub-actions, that can appear in different order, duration, and quantity. These aspects introduce a large intra-class variability, that can be hard to model. Our...

conference paper 2021

document

Model-to-Image Registration via Deep Learning towards Image-Guided Endovascular Interventions

Li, Z. (author), Mancini, Maria Elisabetta (author), Monizzi, Giovanni (author), Andreini, Daniele (author), Ferrigno, Giancarlo (author), Dankelman, J. (author), De Momi, Elena (author)

Cardiologists highlight the need for an intra-operative 3D visualization to assist interventions. The intra-operative 2D X-ray/Digital Subtraction Angiography (DSA) images in the standard clinical workflow limit cardiologists’ views significantly. Compared with image-to-image registration, model-to-image registration is an essential approach...

conference paper 2021

document

On the Importance of Pooling Layer Tuning for Profiling Side-Channel Analysis

Wu, L. (author), Perin, G. (author)

In recent years, the advent of deep neural networks opened new perspectives for security evaluations with side-channel analysis. Profiling attacks now benefit from capabilities offered by convolutional neural networks, such as dimensionality reduction and the inherent ability to reduce the trace desynchronization effects. These neural...

conference paper 2021

document

Probabilistic wind power forecasting combining deep learning architectures

Arends, Eric Lacoa (author), Watson, S.J. (author), Basu, S. (author), Cheneka, B.R. (author)

A series of probabilistic models were bench-marked during the European Energy Markets forecasting Competition 2020 to assess their relative accuracy in predicting aggregated Swedish wind power generation using as input historic weather forecasts from a numerical weather prediction model. In this paper, we report the results of one of these...

conference paper 2020

document

Detecting, classifying, and mapping retail storefronts using street-level imagery

Sharifi Noorian, S. (author), Qiu, S. (author), Psyllidis, A. (author), Bozzon, A. (author), Houben, G.J.P.M. (author)

Up-to-date listings of retail stores and related building functions are challenging and costly to maintain. We introduce a novel method for automatically detecting, geo-locating, and classifying retail stores and related commercial functions, on the basis of storefronts extracted from street-level imagery. Specifically, we present a deep...

conference paper 2020

document

Day-ahead Wind Power Predictions at Regional Scales: Post-processing Operational Weather Forecasts with a Hybrid Neural Network

Basu, S. (author), Watson, S.J. (author), Lacoa Arends, Eric (author), Cheneka, B.R. (author)

A hybrid neural network model, comprising of a convolutional neural network and a multilayer perceptron network, has been developed for day-ahead forecasting of regional scale wind power production. This model requires operational weather forecasts as input and also has the capability to ingest data from ensemble forecasts. Even though the...

conference paper 2020

document

ST-Sem: A Multimodal Method for Points-of-Interest Classification Using Street-Level Imagery

Sharifi Noorian, S. (author), Psyllidis, A. (author), Bozzon, A. (author)

Street-level imagery contains a variety of visual information about the facades of Points of Interest (POIs). In addition to general mor- phological features, signs on the facades of, primarily, business-related POIs could be a valuable source of information about the type and iden- tity of a POI. Recent advancements in computer vision could...

conference paper 2019

document

CNN architectures for GRAPH data

Gama, F. (author), Marques, Antonio G. (author), Leus, G.J.T. (author), Ribeiro, Alejandro (author)

In this ongoing work, we describe several architectures that generalize convolutional neural networks (CNNs) to process signals supported on graphs. The general idea of the replace time invariant filters with graph filters to generate convolutional features and to replace pooling with sampling schemes for graph signals. The different...

conference paper 2019

document

DeepSHM: A deep learning approach for structural health monitoring based on guided Lamb wave technique

Ewald, Vincentius (author), Groves, R.M. (author), Benedictus, R. (author)

In our previous work, we demonstrated how to use inductive bias to infuse a convolutional neural network (CNN) with domain knowledge from fatigue analysis for aircraft visual NDE. We extend this concept to SHM and therefore in this paper, we present a novel framework called DeepSHM which involves data augmentation of captured sensor signals...

conference paper 2019

document

Aggregation Graph Neural Networks

Gama, F. (author), Marques, Antonio G. (author), Ribeiro, Alejandro (author), Leus, G.J.T. (author)

Graph neural networks (GNNs) regularize classical neural networks by exploiting the underlying irregular structure supporting graph data, extending its application to broader data domains. The aggregation GNN presented here is a novel GNN that exploits the fact that the data collected at a single node by means of successive local exchanges...

conference paper 2019

document

Multi-Objective Performance Evaluation of the Detection of Catenary Support Components Using DCNNs

Liu, W. (author), Liu, Zhigang (author), Nunez, Alfredo (author), Wang, Liyou (author), Liu, Kai (author), Lyu, Yang (author), Wang, H. (author)

The goal of this paper is to evaluate from a multi-objective perspective the performance on the detection of catenary support components when using state-of-the-art deep convolutional neural networks (DCNNs). The detection of components is the first step towards a complete automatized monitoring system that will provide actual information about...

conference paper 2018

document

The Role of Articulatory Feature Representation Quality in a Computational Model of Human Spoken-Word Recognition

Scharenborg, O.E. (author), Merkx, Danny (author)

Fine-Tracker is a speech-based model of human speech recognition. While previous work has shown that Fine-Tracker is successful at modelling aspects of human spoken-word recognition, its speech recognition performance is not comparable to that of human performance, possibly due to suboptimal intermediate articulatory feature (AF) representations...

conference paper 2018

document

Sight-seeing in the eyes of deep neural networks

Khademi, S. (author), Shi, X. (author), Mager, Tino (author), Siebes, R.M. (author), Hein, C.M. (author), De Boer, Victor (author), van Gemert, J.C. (author)

We address the interpretability of convolutional neural networks (CNNs) for predicting a geo-location from an image. In a pilot experiment we classify images of Pittsburgh vs Tokyo and visualize the learned CNN filters. We found that varying the CNN architecture leads to variating in the visualized filters. This calls for further...

conference paper 2018