Канал: Song Han

TinyML and Efficient Deep Learning

TinyML and Efficient Deep Learning

TinyML and Efficient Deep Learning on IoT Devices

TinyML and Efficient Deep Learning on IoT Devices

MIT Prof Song Han on TinyML: Reducing AIs Carbon Footprint. Stanford MLSys Seminar Episode 9

MIT Prof Song Han on TinyML: Reducing AIs Carbon Footprint. Stanford MLSys Seminar Episode 9

Efficient AI: Reducing the Carbon Footprint of AI in the Internet of Things (IoT)

Efficient AI: Reducing the Carbon Footprint of AI in the Internet of Things (IoT)

"Once-for-All" DNNs: Simplifying Design of Efficient Models for Diverse Hardware

"Once-for-All" DNNs: Simplifying Design of Efficient Models for Diverse Hardware

Lite Transformer and Hardware-aware Transformer, Invited Talk @ Microsoft Research:

Lite Transformer and Hardware-aware Transformer, Invited Talk @ Microsoft Research:

CVPR'20 tutorial: AutoML for TinyML with Once-for-All Network

CVPR'20 tutorial: AutoML for TinyML with Once-for-All Network

Song Han's PhD Defense. June 1, 2017 @Stanford

Song Han's PhD Defense. June 1, 2017 @Stanford

Song's iPhone 6 plus AirServer Live Stream

Song's iPhone 6 plus AirServer Live Stream

FPGA'17 tutorial

FPGA'17 tutorial

Toward Efficient Deep Neural Network Deployment: Deep Compression and EIE, Song Han

Toward Efficient Deep Neural Network Deployment: Deep Compression and EIE, Song Han

EIE: Efficient Inference Engine on Compressed Deep Neural Network

EIE: Efficient Inference Engine on Compressed Deep Neural Network

ICLR 2016 Best Paper Award: Deep Compression by Song Han

ICLR 2016 Best Paper Award: Deep Compression by Song Han

Techniques for Efficient Implementation of Deep Neural Networks, Song Han @ Embedded vision summit

Techniques for Efficient Implementation of Deep Neural Networks, Song Han @ Embedded vision summit