CSE 240D: Accelerator Design for Deep Learning (Winter 2021)

Instructor: Hadi Esmaeilzadeh

Email: hadi [AT] eng [DOT] ucsd [DOT] edu

Office: CSE 3228

TA:

Joon Kim: jkkim [AT] eng [DOT] ucsd [DOT] edu

Office hours: Thursdays 8-9AM, Zoom

• Syllabus
• Schedule
• Participation
• Presentation
• Critique

Schedule

	Lecture	Date	Presenters
1	Dark Silicon and the End of Multicore Scaling	2/1/2021	Rohan Mahapatra
2	Bit Fusion: Bit-Level Dynamically Composable Architecture for Accelerating Deep Neural Network	2/3/2021	Kevin Mettias
3	Eyeriss: A Spatial Architecture for Energy-Efficient Dataflow for Convolutional Neural Networks	2/8/2021	Arpan Dutta, Jordan Gassaway
4	DNNWEAVER: From High-Level Deep Network Models to FPGA Acceleration	2/10/2021	Ravi Ranjan
5	SCNN: An Accelerator for Compressed-sparse Convolutional Neural Networks	2/17/2021	Edwin Mascarenhas
6	A Domain-Specific Supercomputer for Training Deep Neural Networks	2/22/2021	Pratyush Dwivedi
7	Planaria: Dynamic Architecture Fission for Spatial Multi-Tenant Acceleration of Deep Neural Networks	2/24/2021	Hadi Esmaeilzadeh
8	EIE: Efficient Inference Engine on Compressed Deep Neural Network	3/1/2021	Max Apodaca, Zihao Kong
9	ISAAC: A Convolutional Neural Network Accelerator with In-Situ Analog Arithmetic in Crossbars	3/3/2021	Justin Morris
10	RedEye: Analog ConvNet Image Sensor Architecture for Continuous Mobile Vision	3/8/2021	Yuan Wang
11	RecNMP: Accelerating Personalized Recommendation with Near-Memory Processing	3/10/2021	Jaeyoung Kang