Go To AI Modules

1 Function 2 Reference Model 3 Input/Output Data
4 SubAIMs 5 JSON Metadata 6 Profiles
7 Reference Software 8 Conformance Texting 9 Performance Assessment

1 Functions

The WTI Watermark Embedder (NNW-WWE) provides the Watermarked parameters in response to a given Training dataset and Payload.

2 Reference Model

Figure 1 specifies the Reference Module of the WTI Watermark Embedder (NNW-WWE).

Figure 1 – The WTI Watermark Embedder (NNW-WWE))

3 Input/Output Data

Table 1 specifies the Input and Output Data of the WTI Watermark Embedder (NNW-WWE).

Table 1 – I/O Data of the WTI Watermark Embedder (NNW-WWE)

Input Descriptions
Payload The information to be inserted in the AIM.
Training dataset The dataset used to train the AIM.
Output Descriptions
Watermarked parameters The parameters of the watermarked AIM.

4 SubAIMs

No SubAIMs.

5 JSON Metadata

http://schemas.mpai.community/NNW1/V1.1/AIMs/WTIWatermarkEmbedder.json

6 Profiles

No profiles.

7 Reference Software

8 Conformance Testing

9 Performance Assessment

Go To AI Modules