204

Index

Fast Gradient Sign Method (FGSM), 97

Faster-RCNN, 150

Feed-Forward Network (FFN), 120

FGFI, 177

FPN, 150

FQM, 35

FR-GAL, 151

FullyQT, 121

Fully quantized ViT (Q-ViT), 22

GAL, 151

GELU, 127

Generalized Gauss-Newton matrix (GGN),

105

GIoU, 34

GMM, 78

GOBO, 131

GOT-10K, 14

Gradient Approximation, 3

Grid-GCN, 149

Grid Query (CAGQ), 149

Hessian AWare Quantization (HAWQ), 125

High-Order Residual Quantization

(HORQ), 4

Image Classification, 12

ImageNet, 13

Information Bottleneck (IB), 32

Information Discrepancy-Aware Distillation

for 1-bit Detectors (IDa-Det), 172

Information Rectification Module (IRM), 22

Integer-Only BERT Quantization

(I-BERT), 127

IoU, 150

IR-Net, 84

KL divergence, 110

KR-GAL, 151

LAMB, 27

LayerDrop, 137

Layer-Wise Search for 1-bit Detectors

(LWS-Det), 166

Learned Step Size Quantization (LSQ), 18

LightNN, 8

Local Binary Convolutional Network

(LBCNN), 5, 13

Loss Design, 9

Low-Bit Quantized Detection Transformer

(Q-DETR), 28

Lower Confidence Bound (LCB), 92

LSQ+, 30

M-Filters, 40

Markov Chain Monte Carlo (MCMC), 68

Maximum A posteriori (MAP), 70

Maximum Likelihood Estimation (MLE),

162

Maximum Output Entropy (MOE), 25

MCN Convolution (MCconv), 42

Mean Square Error (MSE), 104

MeliusNet, 7

MetaQuant, 84

Minimum Average Error (MAE), 25

MNIST, 13

MNLI, 126

Modulated Convolutional Networks (MCN),

5

Module-wise Reconstruction Error

Minimization (MREM), 129

MRPC, 135

Multi-Head Attention (MHA), 32

Multi-Head Self-Attention (MHSA), 23

Multi-Layer Perceptron (MLP), 23

Natural Language Processing (NLP), 21

Neural Architecture Search (NAS), 10

Neural networks (NN), 15

Non-Maximum Suppression (NMS), 28

Object Detection and Tracking, 13

Optimization, 10

OTB50, 14

OTB100, 14

Outlier Suppression, 132

PACT, 20

PC-DARTs, 10

PCNNs, 9, 13

POEM, 157

PointNet, 149

PointNet++, 149

Post-training quantization (PTQ), 118

Probability Density Function (PDF), 24

Q-BERT, 125

Q-FC, 32

Q-Linear, 23

QIL, 20

QQP, 128

Quantization, 3

Quantization-aware training (QAT), 21

Quantized neural network (QNN), 16