Results 231 to 240 of about 1,314,055 (250)
Some of the next articles are maybe not open access.
Reliability and Performance Analysis of Logic-in-Memory Based Binarized Neural Networks
IEEE transactions on device and materials reliability, 2021Resistive Random access memory (RRAM) devices together with the material implication (IMPLY) logic are a promising computing scheme for realizing energy efficient reconfigurable computing hardware for edge computing applications.
T. Zanotti, F. Puglisi, P. Pavan
semanticscholar +1 more source
Approximating Binarization in Neural Networks
2019 International Joint Conference on Neural Networks (IJCNN), 2019Binarization of neural networks’ activations may be a requirement for some applications. A typical example is end-to-end learned deep image compression systems where the encoder’s output is requred to be a binary vector. Binarization is non-differentiable, therefore one needs to approximate it in order to train neural networks with stochastic gradient ...
Caglar Aytekin +4 more
openaire +1 more source
ECG signal classification with binarized convolutional neural network
Computers in Biology and Medicine, 2020Arrhythmias are a group of common conditions associated with irregular heart rhythms. Some of these conditions, for instance, atrial fibrillation (AF), might develop into serious syndromes if not treated in time. Therefore, for high-risk patients, early detection of arrhythmias is crucial.
Qing Wu +3 more
openaire +2 more sources
Gradient Matters: Designing Binarized Neural Networks via Enhanced Information-Flow
IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021Binarized neural networks (BNNs) have drawn significant attention in recent years, owing to great potential in reducing computation and storage consumption.
Qi Wang +4 more
semanticscholar +1 more source
An Efficient 6TP SRAM-Based CIM Macro With Column ADCs for Binarized Neural Networks
IEEE Transactions on Circuits and Systems - II - Express BriefsThis brief presents the implementation of a 6T SRAM-based array that computes matrix vector multiplication in binarized neural networks. A 6T SRAM bitcell with PMOS access transistors is proposed, which mitigates the read disturb issue that is attributed
Ikramullah Shah +2 more
semanticscholar +1 more source
Training Binarized Neural Networks Using Ternary Multipliers
IEEE Design & Test, 2021Deep learning offers the promise of intelligent devices that are able to perceive, reason and take intuitive actions. The rising adoption of deep learning techniques has motivated researchers and developers to seek low-cost and high-speed software/hardware solutions for deployment on smart devices.
Amir Ardakani +2 more
openaire +1 more source
Margin-Maximization in Binarized Neural Networks for Optimizing Bit Error Tolerance
Design, Automation and Test in Europe, 2021To overcome the memory wall in neural network (NN) inference systems, recent studies have proposed to use approximate memory, in which the supply voltage and access latency parameters are tuned, for lower energy consumption and faster access at the cost ...
Sebastian Buschjäger +8 more
semanticscholar +1 more source
Binarized Neural Network with Stochastic Memristors
2019 IEEE International Conference on Artificial Intelligence Circuits and Systems (AICAS), 2019This paper proposes the analog hardware implementation of Binarized Neural Network (BNN). Most of the existing hardware implementations of neural networks do not consider the memristor variability issue and its effect on the overall system performance.
Olga Krestinskaya +2 more
openaire +1 more source
Analyzing the Single Event Upset Vulnerability of Binarized Neural Networks on SRAM FPGAs
IEEE International Symposium on Defect and Fault Tolerance in VLSI and Nanotechnology Systems, 2021Neural Networks (NNs) are increasingly used in the last decade in several demanding applications, such as object detection and classification, autonomous driving, etc.
I. Souvatzoglou +4 more
semanticscholar +1 more source
Error-Diffusion Binarization for Neural Networks
1997In optical implementation of neural networks, binarization is often a necessity. Error-diffusion (ED) is presented has a more reliable binarization technic over hardclipping. We use a simple self-organizing learning algorithm for its demonstration.
André Granger +2 more
openaire +1 more source