Related papers: Power efficient ReLU design for neuromorphic compu…
We study neural networks with trainable low-degree rational activation functions and show that they are more expressive and parameter-efficient than modern piecewise-linear and smooth activations such as ELU, LeakyReLU, LogSigmoid, PReLU,…
Activation function is a key component in deep learning that performs non-linear mappings between the inputs and outputs. Rectified Linear Unit (ReLU) has been the most popular activation function across the deep learning community.…
In this paper, we investigate the relationship between deep neural networks (DNN) with rectified linear unit (ReLU) function as the activation function and continuous piecewise linear (CPWL) functions, especially CPWL functions from the…
The spin Hall effect (SHE) generates spin currents within nonmagnetic materials. Previously, studies of the SHE have been motivated primarily to understand its fundamental origin and magnitude. Here we demonstrate, using measurement and…
Recent years have witnessed growing interest in the field of brain-inspired computing based on neural-network architectures. In order to translate the related algorithmic models into powerful, yet energy-efficient cognitive-computing…
The application of the deep learning model in classification plays an important role in the accurate detection of the target objects. However, the accuracy is affected by the activation function in the hidden and output layer. In this…
Activation functions are essential for deep learning methods to learn and perform complex tasks such as image classification. Rectified Linear Unit (ReLU) has been widely used and become the default activation function across the deep…
In past few years, linear rectified unit activation functions have shown its significance in the neural networks, surpassing the performance of sigmoid activations. RELU (Nair & Hinton, 2010), ELU (Clevert et al., 2015), PRELU (He et al.,…
In this paper, we introduce "Power Linear Unit" (PoLU) which increases the nonlinearity capacity of a neural network and thus helps improving its performance. PoLU adopts several advantages of previously proposed activation functions.…
Neural networks (NNs) have been successfully deployed in various fields. In NNs, a large number of multiplyaccumulate (MAC) operations need to be performed. Most existing digital hardware platforms rely on parallel MAC units to accelerate…
In neural network compression, most current methods reduce unnecessary parameters by measuring importance and redundancy. To augment already highly optimized existing solutions, we propose linearity-based compression as a novel way to…
The electric-power dissipation of the anomalous-Hall current injected into a lateral load circuit is studied. The anomalous-Hall current is generated by a $\mathrm{Co_{75}Gd_{25}}$ ferrimagnetic Hall bar and injected into lateral contacts…
Magnetic random-access memory (MRAM) driven by spin-transfer torque (STT) is a major contender for future memory applications. The energy dissipation involved in writing remains problematic, even with the advent of more efficient…
Rectified linear unit (ReLU) activations can also be thought of as 'gates', which, either pass or stop their pre-activation input when they are 'on' (when the pre-activation input is positive) or 'off' (when the pre-activation input is…
Amongst others, the adoption of Rectified Linear Units (ReLUs) is regarded as one of the ingredients of the success of deep learning. ReLU activation has been shown to mitigate the vanishing gradient issue, to encourage sparsity in the…
Neuromorphic models take inspiration from the human brain by adopting bio-plausible neuron models to build alternatives to traditional Machine Learning (ML) and Deep Learning (DL) solutions. The scarce availability of dedicated hardware…
Deep neural networks with rectified linear units (ReLU) are getting more and more popular due to their universal representation power and successful applications. Some theoretical progress regarding the approximation power of deep ReLU…
We investigated the performance of spin transfer torque random access memory (STT-RAM) cell with cross shaped Heusler compound based free layer using micromagnetic simulations. We designed the free layer using Cobalt based Heusler…
Future applications of spin-orbit torque will require new mechanisms to improve the efficiency for switching nanoscale magnetic tunnel junctions (MTJs), while also controlling the magnetic dynamics to achieve fast, nanosecond scale…
Traditional Convolutional Neural Networks (CNNs) typically use the same activation function (usually ReLU) for all neurons with non-linear mapping operations. For example, the deep convolutional architecture Inception-v4 uses ReLU. To…