View source for 18701168. QUANTIZED MACHINE-LEARNING CONFIGURATION INFORMATION (GOOGLE LLC)

=QUANTIZED MACHINE-LEARNING CONFIGURATION INFORMATION=


==Organization Name==

[[:Category:GOOGLE LLC|GOOGLE LLC]]


[[Category:GOOGLE LLC]]

==Inventor(s)==

[[:Category:Jibing Wang of San Jose CA (US)|Jibing Wang of San Jose CA (US)]][[Category:Jibing Wang of San Jose CA (US)]]

[[:Category:Erik Stauffer of Mountain View CA (US)|Erik Stauffer of Mountain View CA (US)]][[Category:Erik Stauffer of Mountain View CA (US)]]

==QUANTIZED MACHINE-LEARNING CONFIGURATION INFORMATION==


This abstract first appeared for US patent application 18701168 titled 'QUANTIZED MACHINE-LEARNING CONFIGURATION INFORMATION




==Original Abstract Submitted==

Aspects describe communicating quantized machine-learning, ML, configuration information over a wireless network. A base station selects () a quantization configuration for quantizing ML configuration information for a deep neural network, DNN, where the quantization configuration indicates one or more quantization formats associated with quantizing the ML configuration information. The base station transmits () an indication of the quantization configuration to a user equipment, UE and transfers (), over the wireless network and with the UE, quantized ML configuration information using the quantization configuration.



[[Category:G06N3/0495]]
[[Category:H04W72/04]]
[[Category:CPC_G06N3/0495]]