A 40nm 4.81TFLOPS/W 8b Floating-Point Training Processor for Non-Sparse Neural Networks Using Shared Exponent Bias and 24-Way Fused Multiply-Add Tree

Jeongwoo Park, SunWoo Lee, Dongsuk Jeon. A 40nm 4.81TFLOPS/W 8b Floating-Point Training Processor for Non-Sparse Neural Networks Using Shared Exponent Bias and 24-Way Fused Multiply-Add Tree. In IEEE International Solid-State Circuits Conference, ISSCC 2021, San Francisco, CA, USA, February 13-22, 2021. pages 148-150, IEEE, 2021. [doi]

@inproceedings{ParkLJ21,
  title = {A 40nm 4.81TFLOPS/W 8b Floating-Point Training Processor for Non-Sparse Neural Networks Using Shared Exponent Bias and 24-Way Fused Multiply-Add Tree},
  author = {Jeongwoo Park and SunWoo Lee and Dongsuk Jeon},
  year = {2021},
  doi = {10.1109/ISSCC42613.2021.9366031},
  url = {https://doi.org/10.1109/ISSCC42613.2021.9366031},
  researchr = {https://researchr.org/publication/ParkLJ21},
  cites = {0},
  citedby = {0},
  pages = {148-150},
  booktitle = {IEEE International Solid-State Circuits Conference, ISSCC 2021, San Francisco, CA, USA, February 13-22, 2021},
  publisher = {IEEE},
  isbn = {978-1-7281-9549-0},
}