Model watermark embedding method and device, computer equipment and storage medium

The invention relates to a model watermark embedding method and device, computer equipment, a storage medium and a computer program product. The method comprises the following steps: acquiring a trained neural network model and to-be-embedded watermark information, and encoding the to-be-embedded watermark information to obtain a watermark vector; determining a to-be-embedded watermark layer from the neural network model, and calculating to obtain embedding parameter position distribution information based on the watermark vector and a network parameter vector corresponding to the to-be-embedded watermark layer; determining a to-be-updated parameter corresponding to each watermark element in the watermark vector from the network parameter vector based on the embedded parameter position distribution information, and updating the corresponding to-be-updated parameter according to each watermark element to obtain each embedded watermark parameter, the position of the to-be-updated parameter determined by embeddi