Zobrazeno 1 - 10
of 84
pro vyhledávání: '"del Barrio, Alberto"'
Memristors offer significant advantages as in-memory computing devices due to their non-volatility, low power consumption, and history-dependent conductivity. These attributes are particularly valuable in the realm of neuromorphic circuits for neural
Externí odkaz:
http://arxiv.org/abs/2407.13410
Crypto-currency markets are known to exhibit inefficiencies, which presents opportunities for profitable cyclic transactions or arbitrage, where one currency is traded for another in a way that results in a net gain without incurring any risk. Quantu
Externí odkaz:
http://arxiv.org/abs/2308.01427
The accuracy requirements in many scientific computing workloads result in the use of double-precision floating-point arithmetic in the execution kernels. Nevertheless, emerging real-number representations, such as posit arithmetic, show promise in d
Externí odkaz:
http://arxiv.org/abs/2305.06946
This paper proposes a training method having multiple cyclic training for achieving enhanced performance in low-bit quantized convolutional neural networks (CNNs). Quantization is a popular method for obtaining lightweight CNNs, where the initializat
Externí odkaz:
http://arxiv.org/abs/2206.12794
Autor:
Mallasén, David, Murillo, Raul, Del Barrio, Alberto A., Botella, Guillermo, Piñuel, Luis, Prieto, Manuel
The posit representation for real numbers is an alternative to the ubiquitous IEEE 754 floating-point standard. In this work, we present PERCIVAL, an application-level posit capable RISC-V core based on CVA6 that can execute all posit instructions, i
Externí odkaz:
http://arxiv.org/abs/2111.15286
Publikováno v:
In Digital Signal Processing May 2024 148
Autor:
Kim, HyunJin, Del Barrio, Alberto A.
Publikováno v:
In Digital Signal Processing March 2024 146
Autor:
Murillo, Raul, Del Barrio, Alberto A., Botella, Guillermo, Kim, Min Soo, Kim, HyunJin, Bagherzadeh, Nader
The Posit Number System was introduced in 2017 as a replacement for floating-point numbers. Since then, the community has explored its application in Neural Network related tasks and produced some unit designs which are still far from being competiti
Externí odkaz:
http://arxiv.org/abs/2102.09262
This paper analyzes the effects of approximate multiplication when performing inferences on deep convolutional neural networks (CNNs). The approximate multiplication can reduce the cost of the underlying circuits so that CNN inferences can be perform
Externí odkaz:
http://arxiv.org/abs/2007.10500
The posit number system is arguably the most promising and discussed topic in Arithmetic nowadays. The recent breakthroughs claimed by the format proposed by John L. Gustafson have put posits in the spotlight. In this work, we first describe an algor
Externí odkaz:
http://arxiv.org/abs/1907.04091