Related papers: Rethinking floating point for deep learning

Multiplication (e.g., convolution) is arguably a cornerstone of modern deep neural networks (DNNs). However, intensive multiplications cause expensive resource costs that challenge DNNs' deployment on resource-constrained edge devices,…

Neural network quantization is widely used to reduce model inference complexity in real-world deployments. However, traditional integer quantization suffers from accuracy degradation when adapting to various dynamic ranges. Recent research…

One of the major promises of quantum computing is the realization of SIMD (single instruction - multiple data) operations using the phenomenon of superposition. Since the dimension of the state space grows exponentially with the number of…

In this paper, the authors propose the idea of a combined integer and floating point multiplier(CIFM) for FPGAs. The authors propose the replacement of existing 18x18 dedicated multipliers in FPGAs with dedicated 24x24 multipliers designed…

This paper introduces HiFloat4 (HiF4), a block floating-point data format tailored for deep learning. Each HiF4 unit packs 64 4-bit elements with 32 bits of shared scaling metadata, averaging 4.5 bits per value. The metadata specifies a…

Quantization is a technique for reducing deep neural networks (DNNs) training and inference times, which is crucial for training in resource constrained environments or applications where inference is time critical. State-of-the-art (SOTA)…

We propose a novel floating-point encoding scheme that builds on prior work involving fixed-point encodings. We encode floating-point numbers using Two's Complement fixed-point mantissas and Two's Complement integral exponents. We used our…

This article presents design techniques proposed for efficient hardware implementation of feedforward artificial neural networks (ANNs) under parallel and time-multiplexed architectures. To reduce their design complexity, after the weights…

Deep Learning Architectures employ heavy computations and bulk of the computational energy is taken up by the convolution operations in the Convolutional Neural Networks. The objective of our proposed work is to reduce the energy…

Neural network training is a memory- and compute-intensive task. Quantization, which enables low-bitwidth formats in training, can significantly mitigate the workload. To reduce quantization error, recent methods have developed new data…

Deep neural network (DNN) quantization converting floating-point (FP) data in the network to integers (INT) is an effective way to shrink the model size for memory saving and simplify the operations for compute acceleration. Recently,…

Embedding Artificial Intelligence onto low-power devices is a challenging task that has been partly overcome with recent advances in machine learning and hardware design. Presently, deep neural networks can be deployed on embedded targets…

Deep learning-based image compression (LIC) has achieved state-of-the-art rate-distortion (RD) performance, yet deploying these models on resource-constrained FPGAs remains a major challenge. This work presents a complete, multi-stage…

Compute-in-memory (CIM) has shown significant potential in efficiently accelerating deep neural networks (DNNs) at the edge, particularly in speeding up quantized models for inference applications. Recently, there has been growing interest…

In this work, we provide energy-efficient architectural support for floating point accuracy. Our goal is to provide accuracy that is far greater than that provided by the processor's hardware floating point unit (FPU). Specifically, for…

Recent advances in convolutional neural networks have considered model complexity and hardware efficiency to enable deployment onto embedded systems and mobile devices. For example, it is now well-known that the arithmetic operations of…

The IEEE 754 floating-point standard is the bedrock of modern computing, but its structural requirements -- a hidden leading bit, Base-2 bit-level normalization, and Sign-Magnitude encoding -- impose significant silicon area and power…

Recent work has shown that 8-bit floating point (FP8) can be used for efficiently training neural networks with reduced computational cost compared to training in FP32/FP16. In this work, we investigate the use of FP8 training in a…

By quantizing network weights and activations to low bitwidth, we can obtain hardware-friendly and energy-efficient networks. However, existing quantization techniques utilizing the straight-through estimator and piecewise constant…

Uniform-precision neural network quantization has gained popularity since it simplifies densely packed arithmetic unit for high computing capability. However, it ignores heterogeneous sensitivity to the impact of quantization errors across…