Neuron-by-Neuron Quantization for Efficient Low-Bit QNN Training
Quantized neural networks (QNNs) are widely used to achieve computationally efficient solutions to recognition problems. Overall, eight-bit QNNs have almost the same accuracy as full-precision networks, but working several times faster. However, the networks with lower quantization levels demonstrat...
Main Authors: | Arlazarov, V.V (Author), Limonova, E. (Author), Nikolaev, D. (Author), Sher, A. (Author), Trusov, A. (Author) |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI
2023
|
Subjects: | |
Online Access: | View Fulltext in Publisher View in Scopus |
Similar Items
-
IR-QNN Framework: An IR Drop-Aware Offline Training of Quantized Crossbar Arrays
by: Mohammed E. Fouda, et al.
Published: (2020-01-01) -
Selection of number of neurons for vector quantization methods
by: Olga Kurasova, et al.
Published: (2008-12-01) -
Training Multi-Bit Quantized and Binarized Networks with a Learnable Symmetric Quantizer
by: Phuoc Pham, et al.
Published: (2021-01-01) -
Gaussian Multiple Access Channels with One-Bit Quantizer at the Receiver †,‡
by: Borzoo Rassouli, et al.
Published: (2018-09-01) -
One-Bit Quantization and Distributed Detection with an Unknown Scale Parameter
by: Fei Gao, et al.
Published: (2015-08-01)