Michael Pfeiffer
YOU?
Author Swipe
View article: How to Digitize Historical Pictorial Collections to Preserve Them in a Digital Life Cycle?
How to Digitize Historical Pictorial Collections to Preserve Them in a Digital Life Cycle? Open
View article: How Many Events Make an Object? Improving Single-frame Object Detection on the 1 Mpx Dataset
How Many Events Make an Object? Improving Single-frame Object Detection on the 1 Mpx Dataset Open
Event cameras are promising novel vision sensors with higher dynamic range and higher temporal resolution compared to frame-based cameras. In contrast to images, single-frame detectors without memory perform poorly on event data. We analyz…
View article: The 6th Albanian Congress of Trauma and Emergency Surgery
The 6th Albanian Congress of Trauma and Emergency Surgery Open
After a three-year quarantine from the deadliest global pandemic of the last century, ASTES is organizing to gather all health professionals in Tirana, The 6th Albanian Congress of Trauma and Emergency Surgery(ACTES 2022) on 11-12 November…
View article: Improving Uncertainty of Deep Learning-based Object Classification on Radar Spectra using Label Smoothing
Improving Uncertainty of Deep Learning-based Object Classification on Radar Spectra using Label Smoothing Open
Object type classification for automotive radar has greatly improved with recent deep learning (DL) solutions, however these developments have mostly focused on the classification accuracy. Before employing DL solutions in safety-critical …
View article: Investigation of Uncertainty of Deep Learning-based Object Classification on Radar Spectra
Investigation of Uncertainty of Deep Learning-based Object Classification on Radar Spectra Open
Deep learning (DL) has recently attracted increasing interest to improve object type classification for automotive radar.In addition to high accuracy, it is crucial for decision making in autonomous vehicles to evaluate the reliability of …
View article: Hybrid SNN-ANN: Energy-Efficient Classification and Object Detection for Event-Based Vision
Hybrid SNN-ANN: Energy-Efficient Classification and Object Detection for Event-Based Vision Open
View article: Bosch Deep Learning Hardware Benchmark
Bosch Deep Learning Hardware Benchmark Open
The widespread use of Deep Learning (DL) applications in science and industry has created a large demand for efficient inference systems. This has resulted in a rapid increase of available Hardware Accelerators (HWAs) making comparison cha…
View article: Multi-Class Uncertainty Calibration via Mutual Information Maximization-based Binning
Multi-Class Uncertainty Calibration via Mutual Information Maximization-based Binning Open
Post-hoc multi-class calibration is a common approach for providing high-quality confidence estimates of deep neural network predictions. Recent work has shown that widely used scaling methods underestimate their calibration error, while a…
View article: Efficient Processing of Spatio-Temporal Data Streams With Spiking Neural Networks
Efficient Processing of Spatio-Temporal Data Streams With Spiking Neural Networks Open
Spiking neural networks (SNNs) are potentially highly efficient models for inference on fully parallel neuromorphic hardware, but existing training methods that convert conventional artificial neural networks (ANNs) into SNNs are unable to…
View article: A generative growth model for thalamocortical axonal branching in primary visual cortex
A generative growth model for thalamocortical axonal branching in primary visual cortex Open
Axonal morphology displays large variability and complexity, yet the canonical regularities of the cortex suggest that such wiring is based on the repeated initiation of a small set of genetically encoded rules. Extracting underlying devel…
View article: Robust Anomaly Detection in Images Using Adversarial Autoencoders
Robust Anomaly Detection in Images Using Adversarial Autoencoders Open
View article: On-manifold Adversarial Data Augmentation Improves Uncertainty Calibration
On-manifold Adversarial Data Augmentation Improves Uncertainty Calibration Open
Uncertainty estimates help to identify ambiguous, novel, or anomalous inputs, but the reliable quantification of uncertainty has proven to be challenging for modern deep networks. In order to improve uncertainty estimation, we propose On-M…
View article: Conditioning by subthreshold synaptic input changes the intrinsic firing pattern of CA3 hippocampal neurons
Conditioning by subthreshold synaptic input changes the intrinsic firing pattern of CA3 hippocampal neurons Open
Unlike synaptic strength, intrinsic excitability is assumed to be a stable property of neurons. For example, learning of somatic conductances is generally not incorporated into computational models, and the discharge pattern of neurons in …
View article: Deep Learning With Spiking Neurons: Opportunities and Challenges
Deep Learning With Spiking Neurons: Opportunities and Challenges Open
Spiking neural networks (SNNs) are inspired by information processing in biology, where sparse and asynchronous binary signals are communicated and processed in a massively parallel fashion. SNNs on neuromorphic hardware exhibit favorable …
View article: Time series anomaly detection based on shapelet learning
Time series anomaly detection based on shapelet learning Open
We consider the problem of learning to detect anomalous time series from an unlabeled data set, possibly contaminated with anomalies in the training data. This scenario is important for applications in medicine, economics, or industrial qu…
View article: Data-driven Summarization of Scientific Articles
Data-driven Summarization of Scientific Articles Open
Data-driven approaches to sequence-to-sequence modelling have been successfully applied to short text summarization of news articles. Such models are typically trained on input-summary pairs consisting of only a single or a few sentences, …
View article: A Generative Growth Model for Thalamocortical Axonal Branching in Primary Visual Cortex
A Generative Growth Model for Thalamocortical Axonal Branching in Primary Visual Cortex Open
Axonal morphology displays large variability and complexity, yet the canonical regularities of the cortex suggest that such wiring is based on the repeated initiation of a small set of genetically encoded rules. Extracting underlying devel…
View article: Data-driven Summarization of Scientific Articles
Data-driven Summarization of Scientific Articles Open
Data-driven approaches to sequence-to-sequence modelling have been successfully applied to short text summarization of news articles. Such models are typically trained on input-summary pairs consisting of only a single or a few sentences, …
View article: Conversion of Continuous-Valued Deep Networks to Efficient Event-Driven Networks for Image Classification
Conversion of Continuous-Valued Deep Networks to Efficient Event-Driven Networks for Image Classification Open
Spiking neural networks (SNNs) can potentially offer an efficient way of doing inference because the neurons in the networks are sparsely activated and computations are event-driven. Previous work showed that simple continuous-value…
View article: Segmentation and classification of colon glands with deep convolutional neural networks and total variation regularization
Segmentation and classification of colon glands with deep convolutional neural networks and total variation regularization Open
Segmentation of histopathology sections is a necessary preprocessing step for digital pathology. Due to the large variability of biological tissue, machine learning techniques have shown superior performance over conventional image process…
View article: Conditioning by Subthreshold Synaptic Input Changes the Intrinsic Firing Pattern of CA3 Hippocampal Neurons
Conditioning by Subthreshold Synaptic Input Changes the Intrinsic Firing Pattern of CA3 Hippocampal Neurons Open
Unlike synaptic strength, intrinsic excitability is assumed to be a stable property of neurons. For example, learning of somatic conductances is generally not incorporated into computational models, and the discharge pattern of neurons in …
View article: Theory and Tools for the Conversion of Analog to Spiking Convolutional Neural Networks
Theory and Tools for the Conversion of Analog to Spiking Convolutional Neural Networks Open
Deep convolutional neural networks (CNNs) have shown great potential for numerous real-world machine learning applications, but performing inference in large CNNs in real-time remains a challenge. We have previously demonstrated that tradi…
View article: Phased LSTM: Accelerating Recurrent Network Training for Long or Event-based Sequences
Phased LSTM: Accelerating Recurrent Network Training for Long or Event-based Sequences Open
Recurrent Neural Networks (RNNs) have become the state-of-the-art choice for extracting patterns from temporal sequences. However, current RNN models are ill-suited to process irregularly sampled data triggered by events generated in conti…
View article: Training Deep Spiking Neural Networks Using Backpropagation
Training Deep Spiking Neural Networks Using Backpropagation Open
Deep spiking neural networks (SNNs) hold the potential for improving the latency and energy efficiency of deep neural networks through data-driven event-based computation. However, training such networks is difficult due to the non-differe…
View article: Deep counter networks for asynchronous event-based processing
Deep counter networks for asynchronous event-based processing Open
Despite their advantages in terms of computational resources, latency, and power consumption, event-based implementations of neural networks have not been able to achieve the same performance figures as their equivalent state-of-the-art de…
View article: Phased LSTM: Accelerating Recurrent Network Training for Long or Event-based Sequences
Phased LSTM: Accelerating Recurrent Network Training for Long or Event-based Sequences Open
Recurrent Neural Networks (RNNs) have become the state-of-the-art choice for extracting patterns from temporal sequences. However, current RNN models are ill-suited to process irregularly sampled data triggered by events generated in conti…
View article: Prediction of Manipulation Actions
Prediction of Manipulation Actions Open
Looking at a person's hands one often can tell what the person is going to do next, how his/her hands are moving and where they will be, because an actor's intentions shape his/her movement kinematics during action execution. Similarly, ac…
View article: Gland segmentation in colon histology images: The glas challenge contest
Gland segmentation in colon histology images: The glas challenge contest Open
Colorectal adenocarcinoma originating in intestinal glandular structures is the most common form of colon cancer. In clinical practice, the morphology of intestinal glands, including architectural appearance and glandular formation, is use…
View article: DVS Benchmark Datasets for Object Tracking, Action Recognition, and Object Recognition
DVS Benchmark Datasets for Object Tracking, Action Recognition, and Object Recognition Open
This data report summarizes a new benchmark dataset in which we converted established visual video benchmarks for object tracking, action recognition and object recognition into spiking neuromorphic datasets, recorded with the DVS output (…
View article: Precise neural network computation with imprecise analog devices
Precise neural network computation with imprecise analog devices Open
The operations used for neural network computation map favorably onto simple analog circuits, which outshine their digital counterparts in terms of compactness and efficiency. Nevertheless, such implementations have been largely supplanted…