Picture for Onur Mutlu

Onur Mutlu

TransPimLib: A Library for Efficient Transcendental Functions on Processing-in-Memory Systems

Apr 23, 2023
Maurus Item, Juan Gómez-Luna, Yuxin Guo, Geraldo F. Oliveira, Mohammad Sadrosadati, Onur Mutlu

Figure 1 for TransPimLib: A Library for Efficient Transcendental Functions on Processing-in-Memory Systems

Figure 2 for TransPimLib: A Library for Efficient Transcendental Functions on Processing-in-Memory Systems

Figure 3 for TransPimLib: A Library for Efficient Transcendental Functions on Processing-in-Memory Systems

Figure 4 for TransPimLib: A Library for Efficient Transcendental Functions on Processing-in-Memory Systems

* Our open-source software is available at https://github.com/CMU-SAFARI/transpimlib

Access Paper or Ask Questions

RedBit: An End-to-End Flexible Framework for Evaluating the Accuracy of Quantized CNNs

Jan 15, 2023
André Santos, João Dinis Ferreira, Onur Mutlu, Gabriel Falcao

Figure 1 for RedBit: An End-to-End Flexible Framework for Evaluating the Accuracy of Quantized CNNs

Figure 2 for RedBit: An End-to-End Flexible Framework for Evaluating the Accuracy of Quantized CNNs

Figure 3 for RedBit: An End-to-End Flexible Framework for Evaluating the Accuracy of Quantized CNNs

Figure 4 for RedBit: An End-to-End Flexible Framework for Evaluating the Accuracy of Quantized CNNs

* 17 pages, 4 figures, 14 tables

Access Paper or Ask Questions

NEON: Enabling Efficient Support for Nonlinear Operations in Resistive RAM-based Neural Network Accelerators

Nov 10, 2022
Aditya Manglik, Minesh Patel, Haiyu Mao, Behzad Salami, Jisung Park, Lois Orosa, Onur Mutlu

Figure 1 for NEON: Enabling Efficient Support for Nonlinear Operations in Resistive RAM-based Neural Network Accelerators

Figure 2 for NEON: Enabling Efficient Support for Nonlinear Operations in Resistive RAM-based Neural Network Accelerators

Figure 3 for NEON: Enabling Efficient Support for Nonlinear Operations in Resistive RAM-based Neural Network Accelerators

Figure 4 for NEON: Enabling Efficient Support for Nonlinear Operations in Resistive RAM-based Neural Network Accelerators

* 11 pages and 11 figures

Access Paper or Ask Questions

Accelerating Neural Network Inference with Processing-in-DRAM: From the Edge to the Cloud

Sep 19, 2022
Geraldo F. Oliveira, Juan Gómez-Luna, Saugata Ghose, Amirali Boroumand, Onur Mutlu

Figure 1 for Accelerating Neural Network Inference with Processing-in-DRAM: From the Edge to the Cloud

Figure 2 for Accelerating Neural Network Inference with Processing-in-DRAM: From the Edge to the Cloud

Figure 3 for Accelerating Neural Network Inference with Processing-in-DRAM: From the Edge to the Cloud

Figure 4 for Accelerating Neural Network Inference with Processing-in-DRAM: From the Edge to the Cloud

* This is an extended and updated version of a paper published in IEEE Micro, pp. 1-14, 29 Aug. 2022. arXiv admin note: text overlap with arXiv:2109.14320

Access Paper or Ask Questions