UGEMM: Unary computing architecture for GEMM applications
General matrix multiplication (GEMM) is universal in various applications, such as signal
processing, machine learning, and computer vision. Conventional GEMM hardware …
processing, machine learning, and computer vision. Conventional GEMM hardware …
Design automation of approximate circuits with runtime reconfigurable accuracy
Leveraging the inherent error tolerance of a vast number of application domains that are
rapidly growing, approximate computing arises as a design alternative to improve the …
rapidly growing, approximate computing arises as a design alternative to improve the …
Approximate softmax functions for energy-efficient deep neural networks
Approximate computing has emerged as a new paradigm that provides power-efficient and
high-performance arithmetic designs by relaxing the stringent requirement of accuracy …
high-performance arithmetic designs by relaxing the stringent requirement of accuracy …
CAxCNN: Towards the use of canonic sign digit based approximation for hardware-friendly convolutional neural networks
The design of hardware-friendly architectures with low computational overhead is desirable
for low latency realization of CNN on resource-constrained embedded platforms. In this …
for low latency realization of CNN on resource-constrained embedded platforms. In this …
Low-Power and Low-Latency Hardware Implementation of Approximate Hyperbolic and Exponential Functions for Embedded System Applications
AM Dalloo, AJ Humaidi, AK Al Mhdawi… - IEEE …, 2024 - ieeexplore.ieee.org
The hyperbolic and exponential functions are widely used in various applications in
engineering fields such as machine learning, Internet of Things (IOT), signal processing, etc …
engineering fields such as machine learning, Internet of Things (IOT), signal processing, etc …
Non-Invasive, Memory Access-Triggered Near-Data Processing for DNN Training Acceleration on GPUs
Currently, GPUs face significant challenges due to limited off-chip bandwidth (BW) and
memory capacity during DNN training. To address these bottlenecks, we propose a memory …
memory capacity during DNN training. To address these bottlenecks, we propose a memory …
Approximate hardware techniques for energy-quality scaling across the system
For error-resilient applications, such as machine learning and signal processing, a
significant improvement in energy efficiency can be achieved by relaxing exactness …
significant improvement in energy efficiency can be achieved by relaxing exactness …
TorchAxf: Enabling Rapid Simulation of Approximate DNN Models Using GPU-Based Floating-Point Computing Framework
This paper presents an approximate floating-point computing framework TorctiAxf1 that
enables fast simulation of various approximate deep neural network (DNN) models …
enables fast simulation of various approximate deep neural network (DNN) models …
Smartphone processor architecture, operations, and functions: current state-of-the-art and future outlook: energy performance trade-off: Energy–performance trade-off …
Balancing energy–performance trade-offs for smartphone processor operations is
undergoing intense research considering the challenges with the evolving technology of …
undergoing intense research considering the challenges with the evolving technology of …
Low Power-Area based Composite 6T-8T SRAM for soft computing applications
S Bharti, A Kumar - 2022 Second International Conference on …, 2022 - ieeexplore.ieee.org
Energy-Time reduction trade-offs for error-tolerant applications has called for an attractive
concept known as “Approximate computing”(AC) at software-hardware levels ex-hibiting …
concept known as “Approximate computing”(AC) at software-hardware levels ex-hibiting …