Progressive growing autoencoder

An apparatus comprising:

• at least one processor;
• and
• at least one memory storing instructions, that when executed by the at least one processor, cause the apparatus to perform:
• obtaining training data;
• obtaining a plurality of compression rates;
• and
• training compression blocks of an encoder of an autoencoder and reconstruction blocks of a decoder of the autoencoder progressively to provide a progressive growing autoencoder based on the training data by training a first block of the encoder and a last block of the decoder to provide a first compression rate of the plurality of compression rates and then subsequently progressively training remaining compression blocks of the encoder and remaining reconstruction blocks of the decoder to provide respective remaining compression rates of the plurality of compression rates.

A method comprising:

• obtaining training data;
• obtaining a plurality of compression rates;
• and
• training compression blocks of an encoder of an autoencoder and reconstruction blocks of a decoder of the autoencoder progressively to provide a progressive growing autoencoder based on the training data by training a first block of the encoder and a last block of the decoder to provide a first compression rate of the plurality of compression rates and then subsequently progressively training remaining compression blocks of the encoder and remaining reconstruction blocks of the decoder to provide respective remaining compression rates of the plurality of compression rates.

An apparatus comprising:

• at least one processor;
• and
• at least one memory storing instructions, that when executed by the at least one processor, cause the apparatus to perform:
• receiving a model structure of an encoder, trained weights of the encoder, a plurality of compression rates and a required number of compression blocks for execution at each compression rate;
• receiving, from a base station, an indication of a compression rate to be used;
• and
• setting an encoder model of the encoder based on the model structure of the encoder, the trained weights of the encoder, the plurality of compression rates and the required number of compression blocks for execution at each compression rate.