報(bào)告題目：Deep Learning Systems: Design and Implementation（一）（二）（三）

報(bào)告時(shí)間：2023年7月4、5、7日下午15:30

報(bào)告地點(diǎn)：437bwin必贏國(guó)際官網(wǎng)B404會(huì)議室

報(bào)告人：楊東林

報(bào)告人國(guó)籍：中國(guó)

報(bào)告人單位：美國(guó)英偉達(dá)公司（NVIDIA）

報(bào)告人簡(jiǎn)介：楊東林，美國(guó)英偉達(dá)公司高級(jí)工程師，主要研究深度學(xué)習(xí)框架、深度學(xué)習(xí)編譯器相關(guān)的課題。于2022年在美國(guó)University of North Carolina at Charlotte 獲得博士學(xué)位。博士期間研究方向主要包括大規(guī)模分布式機(jī)器學(xué)習(xí)，深度學(xué)習(xí)資源調(diào)度等課題。在博士期間曾在HPDC, Middleware等國(guó)際會(huì)議發(fā)表論文。

報(bào)告摘要：Deep learning methods have revolutionized a number of fields in Artificial Intelligence and Machine Learning in recent years. The widespread adoption of deep learning methods have in no small part been driven by the widespread availability of easy-to-use deep learning systems, such as PyTorch and TensorFlow. The goal of this talk is to understand the working of DL framework under the hood. This talk provides an understanding and overview of the “full stack” of deep learning systems, ranging from the high-level modeling design of modern deep learning systems, to the basic implementation of automatic differentiation tools, to the distributed implementation of efficient algorithms, to the compilation of deep learning models into deployable modules.

邀請(qǐng)人：程大釗