LNS-Madam: Low-Precision Training in Logarithmic Number System using Multiplicative Weight Update
Zhao, Jiawei, Dai, Steve, Venkatesan, Rangharajan, Zimmer, Brian, Ali, Mustafa, Liu, Ming-Yu, Khailany, Brucek, Dally, Bill, Anandkumar, Anima
–arXiv.org Artificial Intelligence
Representing deep neural networks (DNNs) in low-precision is a promising approach to enable efficient acceleration and memory reduction. Previous methods that train DNNs in low-precision typically keep a copy of weights in high-precision during the weight updates. Directly training with low-precision weights leads to accuracy degradation due to complex interactions between the low-precision number systems and the learning algorithms. To address this issue, we develop a co-designed low-precision training framework, termed LNS-Madam, in which we jointly design a logarithmic number system (LNS) and a multiplicative weight update algorithm (Madam). We prove that LNS-Madam results in low quantization error during weight updates, leading to stable performance even if the precision is limited. We further propose a hardware design of LNS-Madam that resolves practical challenges in implementing an efficient datapath for LNS computations. Our implementation effectively reduces energy overhead incurred by LNS-to-integer conversion and partial sum accumulation. Experimental results show that LNS-Madam achieves comparable accuracy to full-precision counterparts with only 8 bits on popular computer vision and natural language tasks. Compared to FP32 and FP8, LNS-Madam reduces the energy consumption by over 90% and 55%, respectively.
arXiv.org Artificial Intelligence
Aug-23-2022
- Country:
- North America
- United States
- Michigan (0.04)
- Texas > Travis County
- Austin (0.04)
- Minnesota > Hennepin County
- Minneapolis (0.14)
- Nevada > Clark County
- Las Vegas (0.04)
- Maryland > Prince George's County
- College Park (0.04)
- Louisiana > Orleans Parish
- New Orleans (0.04)
- Utah > Salt Lake County
- Salt Lake City (0.04)
- Florida > Miami-Dade County
- Miami (0.04)
- Washington > King County
- Seattle (0.04)
- California
- Yolo County > Davis (0.04)
- Orange County > Irvine (0.04)
- Alameda County > Berkeley (0.04)
- Santa Clara County
- Santa Clara (0.14)
- Stanford (0.04)
- Los Angeles County
- Pasadena (0.04)
- Los Angeles (0.04)
- New York > New York County
- New York City (0.04)
- Canada
- Ontario > Toronto (0.14)
- Quebec > Montreal (0.04)
- British Columbia > Metro Vancouver Regional District
- Vancouver (0.04)
- United States
- Europe
- Austria (0.04)
- Spain
- Valencian Community > Valencia Province
- Valencia (0.04)
- Catalonia > Barcelona Province
- Barcelona (0.04)
- Valencian Community > Valencia Province
- Italy > Calabria
- Catanzaro Province > Catanzaro (0.04)
- France > Hauts-de-France
- Asia > China
- Jiangsu Province > Nanjing (0.04)
- Africa > Middle East
- Egypt > Cairo Governorate > Cairo (0.04)
- North America
- Genre:
- Research Report
- New Finding (0.48)
- Promising Solution (0.34)
- Research Report
- Industry:
- Education (0.68)
- Semiconductors & Electronics (0.46)
- Technology: