Related papers: Development of Deep Neural Network First-Level Har…
To speedup Deep Neural Networks (DNN) accelerator design and enable effective implementation, we propose HybridDNN, a framework for building high-performance hybrid DNN accelerators and delivering FPGA-based hardware implementations. Novel…
Real-time Deep Neural Network (DNN) inference with low-latency requirement has become increasingly important for numerous applications in both cloud computing (e.g., Apple's Siri) and edge computing (e.g., Google/Waymo's driverless car).…
Recently spiking neural networks (SNNs), the third-generation of neural networks has shown remarkable capabilities of energy-efficient computing, which is a promising alternative for deep neural networks (DNNs) with high energy consumption.…
The reconstruction of charged particle trajectories in tracking detectors is a key problem in the analysis of experimental data for high-energy and nuclear physics. The amount of data in modern experiments is so large that classical…
The physics goals the Belle II experiment require an exceptionally good alignment of all the components of the Belle II tracker. The Belle II tracker is composed of the DEPFET based pixel silicon detector, four layers of double sided…
We developed task-optimized deep neural networks (DNNs) that achieved state-of-the-art performance in different evaluation scenarios for automatic music tagging. These DNNs were subsequently used to probe the neural representations of…
Spiking neural networks (SNNs) recently gained momentum due to their low-power multiplication-free computing and the closer resemblance of biological processes in the nervous system of humans. However, SNNs require very long spike trains…
Many techniques have been developed, such as model compression, to make Deep Neural Networks (DNNs) inference more efficiently. Nevertheless, DNNs still lack excellent run-time dynamic inference capability to enable users trade-off accuracy…
The Belle detector operated at KEKb B-factory in 1999-2010 was one of the most remarkable experiments in the field of elementary particle physics of the last decades. The Belle successor, Belle II collaboration, is aimed to operate the…
Current long-baseline neutrino-oscillation experiments such as NO$\nu$A and T2K are mainly sensitive to physics in the neighbourhood of the first oscillation maximum of the $\nu_\mu \to \nu_e$ oscillation probability. The future Deep…
Deep neural networks (DNNs) have substantial computational requirements, which greatly limit their performance in resource-constrained environments. Recently, there are increasing efforts on optical neural networks and optical computing…
With the rapid development of deep learning, Deep Spiking Neural Networks (DSNNs) have emerged as promising due to their unique spike event processing and asynchronous computation. When deployed on neuromorphic chips, DSNNs offer…
The Belle II experiment at the SuperKEKB accelerator is a next-generation B-factory aiming to collect 50 ab$^{-1}$, about 50 times the data collected at Belle, to study rare processes and make precision measurements that may expose physics…
Deep neural network (DNN) inference relies increasingly on specialized hardware for high computational efficiency. This work introduces a field-programmable gate array (FPGA)-based dynamically configurable accelerator featuring systolic…
A deep neural network (DNN) that can reliably model muscle responses from corresponding brain stimulation has the potential to increase knowledge of coordinated motor control for numerous basic science and applied use cases. Such cases…
Dynamic Graph Neural Networks (DGNNs) are becoming increasingly popular due to their effectiveness in analyzing and predicting the evolution of complex interconnected graph-based systems. However, hardware deployment of DGNNs still remains…
Deep neural networks (DNN) are increasingly being used to learn controllers due to their excellent approximation capabilities. However, their black-box nature poses significant challenges to closed-loop stability guarantees and performance…
Deep Neural Networks (DNNs) are increasingly deployed across distributed and resource-constrained platforms, such as System-on-Chip (SoC) accelerators and edge-cloud systems. DNNs are often partitioned and executed across heterogeneous…
The upgrade of the track classification and selection step of the CMS tracking to a Deep Neural Network is presented. The CMS tracking follows an iterative approach: tracks are reconstructed in multiple passes starting from the ones that…
This article describes the Silicon Track Trigger (STT) which has been fully commissioned in 2004 at the DO experiment. The STT allows to enrich already at the second trigger level the data sample with events containing B-mesons. The STT…