Search Results for author:
Found 6 papers, 3 papers with code
Post-Training Sparsity-Aware Quantization
Moreover, instead of quantizing activation-by-activation to 4 bits, we focus on pairs of 8-bit activations and examine whether one of the two is equal to zero.
Post-Training BatchNorm Recalibration
However, the method of accommodating more than one thread in a shared MAC unit may contribute noise to the computations, thereby changing the internal statistics of the model.
Non-Blocking Simultaneous Multithreading: Embracing the Resiliency of Deep Neural Networks
Inspired by conventional CPU simultaneous multithreading (SMT) that increases computer resource utilization by sharing them across several threads, we propose non-blocking SMT (NB-SMT) designated for DNN accelerators.
Robust Quantization: One Model to Rule Them All
Neural network quantization methods often involve simulating the quantization process during training, making the trained model highly dependent on the target bit-width and precise way quantization is performed.
Thanks for Nothing: Predicting Zero-Valued Activations with Lightweight Convolutional Neural Networks
Convolutional neural networks (CNNs) introduce state-of-the-art results for various tasks with the price of high computational demands.
Spatial Correlation and Value Prediction in Convolutional Neural Networks
Convolutional neural networks (CNNs) are a widely used form of deep neural networks, introducing state-of-the-art results for different problems such as image classification, computer vision tasks, and speech recognition.
