notesum.ai

Published at December 4

Enhancing Supply Chain Visibility with Generative AI: An Exploratory Case Study on Relationship Prediction in Knowledge Graphs

cs.CE

cs.AI

Released Date: December 4, 2024

Authors: Ge Zheng¹, Alexandra Brintrup²

Aff.: ¹Institute for Manufacturing, University of Cambridge, United Kingdom; ²Alan Turing Institute, London, United Kingdom

Arxiv: http://arxiv.org/pdf/2412.03390v1

Refer to caption

ANN		CNN1D		LogReg		LSTM		AutoEncoder
Layer Name	Param -eters	Layer Name	Param -eters	Layer Name	Param -eters	Layer Name	Param -eters	Layer Name	Param -eters
Linear¹	300	Conv1D¹	32,(7),2	Linear¹	200	LSTM¹	16,16,bi	Linear¹ (Encoder)	96
BatchNorm¹	300	ReLU¹	–	–	–	–	–	ReLU¹ (Encoder)	–
ReLU¹	–	AvgPooling¹	-,(7),2	–	–	–	–	Linear² (Encoder)	48
–	–	BatchNorm¹	32	–	–	–	–	ReLU² (Encoder)	–
Linear²	300	Conv1D²	64,(7),1	Linear²	2	LSTM²	16,16,bi	Linear¹ (Decoder)	48
BatchNorm²	300	ReLU²	–	–	–	–	–	ReLU¹ (Decoder)	–
ReLU²	–	AvgPooling²	-,(7),1	–	–	–	–	Linear² (Decoder)	96
–	–	BatchNorm²	64	–	–	–	–	ReLU² (Decoder)	–
Linear³	300	CNN-1D³	64,(7),1	–	–	–	–	–	–
BatchNorm³	300	ReLU³	–	–	–	–	–	–	–
ReLU³	–	AvgPooling³	-,(7),1	–	–	–	–	–	–
–	–	BatchNorm³	64	–	–	–	–	–	–
–	–	Flatten	–	–	–	–	–	–	–
FC	2	FC	2	–	–	FC	2	FC	2
Softmax	–	Softmax	–	Sigmoid	–	Softmax	–	Softmax	–