How to update the memory in the network?

donggong1 / memae-anomaly-detection

MemAE for anomaly detection. -- Gong, Dong, et al. "Memorizing Normality to Detect Anomaly: Memory-augmented Deep Autoencoder for Unsupervised Anomaly Detection". ICCV 2019.

https://donggong1.github.io/anomdec-memae.html

MIT License

465 stars 103 forks source link

How to update the memory in the network? #17

Closed NeuZhangQiang closed 3 years ago

NeuZhangQiang commented 3 years ago

The paper said:

During training, the memory `M` is updated through optimization via backpropagation and gradient descent.

Does it mean that M is the optimized parameters in the network?

Thank you very much for any suggestion.

donggong1 commented 3 years ago

Yes. You can understand it in this way. In some view, M can be seen as a compression of the prototypical dataset in the normal data.

NeuZhangQiang commented 3 years ago

Thank you very much !