Protecting Deep Neural Network Intellectual Property with Architecture-Agnostic Input Obfuscation.
Brooks OlneyRobert KaramPublished in: ACM Great Lakes Symposium on VLSI (2022)
Keyphrases
- intellectual property
- neural network
- network architecture
- desired output
- clef ip
- management system
- patent search
- associative memory
- patent information
- software architecture
- neural architecture
- real time
- reverse engineering
- multi layer
- access control
- decision making
- neural network is trained
- case study
- artificial neural networks
- neural network model
- input data
- multi layer perceptron
- fuzzy logic
- prior art
- patent documents
- hidden layer
- metadata
- radial basis function
- back propagation